A hybrid intelligent-based education resource big data mining method and system

Abstract

The application discloses a kind of education resource big data mining method and system based on hybrid intelligence, belongs to education resource mining technical field, method includes: according to the various potential education analysis needs of user information determination has, each potential education analysis needs is filtered, obtains each target education analysis need;Obtain the education data information of user, according to each target education analysis need of education data information collaborative evaluation, obtains the collaborative evaluation result of each target education analysis need, and the collaborative evaluation result is marked as collaborative analysis need for the target education analysis need needing collaborative analysis;The collaborative analysis needs of each user are analyzed, and the collaborative data user and collaborative data range corresponding to each user are obtained;According to the collaborative data user and collaborative data range of each user, collaborative processing is carried out.

Description

Technical Field

[0001] This invention belongs to the field of educational resource mining technology, specifically a method and system for mining educational resource big data based on hybrid intelligence. Background Technology

[0002] Current data mining in the education field primarily relies on teaching data accumulated within individual institutions, such as school-built student registration management systems and user behavior logs from online education platforms. This data model has significant limitations: First, the single data dimension leads to biased analysis results; for example, a university can only optimize its teaching strategies based on its own course data, unable to compare the teaching effectiveness of similar courses in other institutions. Second, limited data scale affects the model's generalization ability; for example, the student sample size of regional training institutions is insufficient to support nationwide learning predictions. Third, redundant development leads to resource waste; different institutions independently develop similar analysis tools, resulting in fragmented technological investment and inconsistent standards. While existing technologies attempt to achieve cross-institutional analysis through centralized data storage, this requires transmitting raw data to a central server, raising the risk of leakage of core teaching data, such as sensitive information like student grades and teacher evaluations, which may be illegally obtained or reused.

[0003] In order to solve the above problems and achieve deeper mining of educational data, this invention provides a method and system for mining big data of educational resources based on hybrid intelligence. Summary of the Invention

[0004] To address the problems of the above solutions, this invention provides a method and system for mining educational resource big data based on hybrid intelligence.

[0005] The objective of this invention can be achieved through the following technical solutions: A big data mining system for educational resources based on hybrid intelligence, comprising a platform and a user terminal; Furthermore, the platform and each user terminal establish a communication connection.

[0006] The user terminal includes a resource analysis module; The resource analysis module is used to perform needs analysis on users, determine various potential educational analysis needs based on user information, filter each potential educational analysis need, and obtain each target educational analysis need. The system acquires users' educational data and conducts collaborative evaluations of various target educational analysis needs based on the educational data. It obtains the collaborative evaluation results for each target educational analysis need, including those that do not require collaborative analysis and those that do. The system marks the target educational analysis needs that require collaborative analysis as collaborative analysis needs and sends the educational data and collaborative analysis needs to the platform.

[0007] Furthermore, based on user information, various potential educational analytics needs are identified, including: The platform establishes a reserve database, which is used to store various educational analysis needs and the applicable conditions corresponding to each educational analysis need. Establish a matching model, the expression of which is: ; In the formula: (s i ,YU) is the input data, s i This indicates the applicable conditions for the corresponding educational analysis needs in the reserve, where i represents the corresponding educational need, i = 1, 2, ..., n, and n is the number of educational analysis needs in the reserve; YU represents user information; s i →YU indicates that the applicable conditions match the user information; the output data is the matching value PU(s) i (YU), matching value is 1 or 0; The user information and the applicable conditions of the corresponding educational analysis needs in the reserve are integrated into the input data and input into the matching model for analysis to obtain the corresponding matching value; Each educational analytics requirement with a matching value of 1 is marked as a potential educational analytics requirement.

[0008] Furthermore, the various potential educational analysis needs are screened, including: The system presents all potential educational analytics needs to users, who then filter and mark the remaining potential needs as target educational analytics needs.

[0009] Furthermore, the various potential educational analysis needs are screened, including: The process involves obtaining the user's target requirement standards, collecting features from each potential educational analysis need based on these standards, and acquiring the target requirement features corresponding to each potential educational analysis need. These features are either related to the target requirements or are calculated or derived from them. The target requirement features of each potential educational analysis need are then calibrated using the target requirement standards, and the calibrated potential educational analysis needs are marked as target educational analysis needs.

[0010] Furthermore, based on educational data, a collaborative assessment of the educational analysis needs for each objective is conducted, including: Identify the data requirements for each target educational analysis need, and determine whether the user's educational data meets the data requirements for each target educational analysis need based on the educational data information. Collaborative assessment results that do not meet the data requirements for the target education analysis are marked as requiring collaborative analysis; The platform updates the corresponding effectiveness standards for each educational analysis need in the reserve in real time; it estimates the expected effectiveness of the target educational analysis needs that meet the data requirements based on educational data information; it matches each effectiveness standard from the reserve according to the target educational analysis needs; it compares each effectiveness standard with the corresponding estimated effectiveness to obtain the effectiveness comparison results, which include the effectiveness being qualified and the effectiveness being unqualified. Mark the results of the collaborative assessment of the needs for analysis of the target education that do not meet the requirements as requiring collaborative analysis; Mark the results of the collaborative assessment of the target education needs analysis that meet the requirements as not requiring collaborative analysis.

[0011] Furthermore, comparisons are made between each performance criterion and its corresponding predicted performance, including: Calculate the optimal value between the standard effect and the corresponding predicted effect, and label the obtained optimal value as YR. j j represents the corresponding effect standard, j = 1, 2, ..., m, where m is the number of effect standards; The comprehensive comparison value is calculated using the following formula: ; In the formula: UB is the comprehensive comparison value; b j This represents the corresponding proportionality coefficient, with a value range of (0, 1). When the overall comparison value is greater than the threshold X1, the evaluation effect is unqualified; The evaluation effect is qualified when the comprehensive comparison value is not greater than the threshold X1.

[0012] The platform includes a resource information database, a demand analysis module, and a collaborative processing module. The resource information database is used to store educational data information of each user and platform resource data.

[0013] The requirements analysis module analyzes the collaborative analysis requirements of each user to obtain the collaborative data users and the scope of collaborative data for each user.

[0014] Furthermore, the collaborative analysis needs of each user are analyzed, including: Obtain the corresponding effect criteria for each collaborative analysis requirement, and determine the requirement data information for the corresponding collaborative analysis requirement based on each effect criterion. Each collaborative analysis requirement and requirement data is input into the resource information database for matching and analysis to obtain candidate collaborative solutions that meet each collaborative analysis requirement. The candidate collaborative solutions are then filtered to obtain the target collaborative solution. Based on the target collaborative solution, the corresponding collaborative data users and the scope of collaborative data are determined.

[0015] The collaborative processing module is used to perform collaborative processing based on the collaborative data users and the scope of collaborative data of each user, and to generate corresponding platform resource data according to the collaborative processing process, and send the platform resource data to the resource information database for storage.

[0016] A method for mining educational resource big data based on hybrid intelligence, the method includes: Based on user information, identify various potential educational analysis needs, filter these potential educational analysis needs, and obtain the target educational analysis needs. Acquire users' educational data information, conduct collaborative evaluation of various target educational analysis needs based on the educational data information, obtain the collaborative evaluation results of various target educational analysis needs, and mark the target educational analysis needs that require collaborative analysis according to the collaborative evaluation results as collaborative analysis needs.

[0017] Analyze the collaborative analysis needs of each user to obtain the corresponding collaborative data users and collaborative data range for each user; and perform collaborative processing based on the collaborative data users and collaborative data range for each user.

[0018] Compared with the prior art, the beneficial effects of the present invention are: Through the coordinated efforts of its various modules, the system significantly improves the efficiency and value of educational resource circulation. By constructing a secure and reliable collaborative data mining mechanism, it enables deep integration of cross-institutional educational data while protecting the data sovereignty of all parties. The system supports educational institutions in collaborative data applications without needing to share raw data, thereby increasing the utilization rate and value of high-quality teaching resources. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a block diagram illustrating the principle of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] like Figure 1 As shown, an educational resource big data mining system based on hybrid intelligence includes a platform and a user terminal. The communication connection between the platform and each user terminal can also be achieved through other methods to transmit data.

[0023] The platform can be configured based on the cloud or a local server; the user end is used by users who have educational resources or need to utilize educational resources.

[0024] The user terminal includes a resource analysis module, The resource analysis module is used to perform needs analysis on users. Based on user information, it identifies various potential educational analysis needs. User information includes an introduction to the user and education-related information, such as school information if the user is a school, teacher information if the user is a teacher, and other education-related information such as courses and teaching methods. Users fill in the information according to a preset template. The module then displays each potential educational analysis need to the user, who then selects each target educational analysis need based on their actual educational situation. If no selection is made, all educational analysis needs can be considered as target educational analysis needs. Alternatively, users can preset target requirements such as analysis cost and expected results to filter the potential educational analysis needs, marking the remaining potential educational analysis needs as target educational analysis needs.

[0025] The system acquires users' educational data to summarize various educational data and related information such as data volume. Based on the educational data, it evaluates each target educational analysis need to determine whether each target educational analysis need requires collaborative analysis using other users' educational data. It obtains corresponding collaborative evaluation results, which include those that do not require collaborative analysis and those that do. Target educational analysis needs that require collaborative analysis are marked as collaborative analysis needs. The educational data and collaborative analysis needs are then sent to the platform. The sent educational data does not contain specific educational data or educational data that the user is prohibited from disclosing.

[0026] In one embodiment, various potential educational analysis needs of users can be determined based on existing technology and user information, that is, various needs for analyzing educational data that users may have, and then corresponding analysis results can be obtained, such as potential educational analysis needs such as student mental health assessment, student learning status assessment, and teacher teaching quality assessment. For example, user information can be matched with various potential educational analysis needs preset by the platform; matching can be performed based on user similarity, etc.

[0027] In one embodiment, various potential educational analytics needs are identified based on user information, including: The platform establishes a reserve database to store various educational analysis needs for mining and analyzing educational data, as well as the applicable conditions for these needs, i.e., under what circumstances would a user have such an educational analysis need. The platform compiles and updates this database based on relevant historical data.

[0028] Establish a matching model, the expression of which is: ; In the formula: (s i ,YU) is the input data, s i This indicates the applicable conditions for the corresponding educational analysis needs in the reserve, where i represents the corresponding educational need, i = 1, 2, ..., n, and n is the number of educational analysis needs in the reserve; YU represents user information; s i →YU indicates that the applicable conditions match the user information; the output data is the matching value PU(s) i , YU), and the staff will train according to the corresponding training set marked in the reserve; The user information and the applicable conditions of the corresponding educational analysis needs in the reserve are integrated into the input data and input into the matching model for analysis to obtain the corresponding matching value; Each educational analytics requirement with a matching value of 1 is marked as a potential educational analytics requirement.

[0029] In one embodiment, collaborative evaluation is performed on various target educational analysis needs based on educational data information. The evaluation is based on whether the existing user's educational data can meet the data requirements of the target educational analysis needs. If it can, it is considered that collaborative analysis is not required; otherwise, it is considered that collaborative analysis is required. The evaluation is based on the data requirements for achieving the target educational analysis needs.

[0030] In one embodiment, this embodiment further evaluates the previous embodiment, considering the best achievable analytical effect, and collaboratively evaluates the various target educational analysis needs based on educational data information, including: Identify the data requirements for each target educational analysis need. Data requirements are the types and amounts of data needed to achieve the target educational analysis need. For example, to build a student status prediction model based on educational data, the corresponding types and amounts of data are needed to set up a training set for training. Determine whether the user's educational data meets the data requirements for each target educational analysis need based on the educational data information. Collaborative assessment results that do not meet the data requirements for the target education analysis are marked as requiring collaborative analysis; The platform updates the various effectiveness standards corresponding to each educational analysis need in the reserve in real time, such as analysis accuracy and analysis efficiency. The platform determines the effectiveness standards based on the current achievable effects of the educational analysis need. For example, the best achievable effect can be used as the effectiveness standard. The specific settings are determined by the platform. The best overall effectiveness standards can also be determined by factors such as cost. Based on educational data information, the estimated effect of meeting the target educational analysis needs is predicted. The prediction is made based on existing prediction techniques, such as predicting the effect of achieving the target transaction analysis needs according to existing educational data based on existing historical data, and thus forming the predicted effect. Based on the needs of the target education analysis, various effectiveness standards are matched from the reserve, and each effectiveness standard is compared with the corresponding estimated effect to obtain the effectiveness comparison results, which include the effectiveness being qualified and the effectiveness being unqualified. Mark the results of the collaborative assessment of the needs for analysis of the target education that do not meet the requirements as requiring collaborative analysis; Mark the results of the collaborative assessment of the target education needs analysis that meet the requirements as not requiring collaborative analysis.

[0031] In one embodiment, the effect is compared with the corresponding estimated effect according to each effect standard. Specifically, the comparison is made according to the user's needs. For example, if any estimated effect fails to meet the effect standard, the effect is deemed unqualified. Alternatively, an allowable difference range for each effect can be preset. As long as the effect is within the difference range, the effect is deemed qualified.

[0032] In one embodiment, a comparison is made between each performance criterion and the corresponding predicted performance, including: Calculate the optimal performance value between the performance standard and the corresponding predicted performance, i.e., the proportion by which the performance standard is worse than the predicted performance. For example, if the performance standard is 95% analytical accuracy and the predicted performance is 80% analytical accuracy, then the optimal performance value = (95-80) ÷ 80; label the obtained optimal performance value as YR. j j represents the corresponding effect standard, j = 1, 2, ..., m, where m is the number of effect standards; The comprehensive comparison value is calculated using the following formula: ; In the formula: UB is the comprehensive comparison value; b j This represents the corresponding proportionality coefficient, with a value range of (0, 1). When the overall comparison value is greater than the threshold X1, the evaluation effect is unqualified; The evaluation effect is qualified when the comprehensive comparison value is not greater than the threshold X1.

[0033] The threshold X1 can be set according to user needs. For example, a user can select one case as the standard to calculate the comprehensive comparison value and use the comprehensive comparison value as the threshold X1. Alternatively, multiple cases can be selected for comprehensive determination of the threshold X1.

[0034] The platform includes a resource information database, a demand analysis module, and a collaborative processing module. The resource information database is used to store educational data information of each user and platform resource data.

[0035] The requirements analysis module analyzes the collaborative analysis requirements of each user to obtain the collaborative data users and the scope of collaborative data for each user.

[0036] In one embodiment, the collaborative analysis needs of various users are analyzed, including: Obtain the various effect standards corresponding to each collaborative analysis requirement, and determine the required data information for the corresponding collaborative analysis requirement based on each effect standard, that is, the relevant information such as the types and amounts of data required to achieve each effect standard. Each collaborative analysis requirement and requirement data is input into the resource information database for matching and analysis to obtain candidate collaborative solutions that meet each collaborative analysis requirement. The candidate collaborative solutions are then filtered to obtain the target collaborative solution. Based on the target collaborative solution, the corresponding collaborative data users and the scope of collaborative data are determined.

[0037] In one embodiment, various collaborative analysis needs and their data are input into a resource information database for matching analysis. This matching analysis can be performed using existing methods, where platform resource data and various educational data jointly match the collaborative analysis needs and their data. The goal is to find combinations of data from each user and the platform that meet the matching requirements. Each combination forms a candidate collaborative solution, with the matching requirement being the fulfillment of the corresponding data for each collaborative analysis need. For platform resource data, resources such as model parameters with direct results are considered to meet the matching requirements if they address the corresponding collaborative analysis needs. Intelligent analysis can be performed using technologies such as machine learning to generate candidate collaborative solutions. The platform is also considered a collaborative data user.

[0038] In one embodiment, the selection of collaborative solutions can be prioritized based on factors such as cost, overall effectiveness, and collaborative efficiency, with the highest-priority collaborative solution being marked as the target collaborative solution. Alternatively, the platform can assist users in selecting the target collaborative solution that satisfies them.

[0039] The collaborative processing module is used to perform collaborative processing based on the collaborative data users and the scope of collaborative data of each user, so as to realize the user's various target education analysis needs; and to generate corresponding platform resource data according to the collaborative processing process, and send the platform resource data to the resource information database for storage.

[0040] For example, based on various existing methods such as federated learning and adaptive differential privacy protection, educational data sharing applications can be implemented within the scope of collaborative data to meet users' various target educational analysis needs.

[0041] For example, the platform initializes a global model based on collaborative analysis needs, distributes copies of the global model to various collaborative data users and users, and the users and collaborative data users train the global model copy using local data to form corresponding model parameters. The model parameters are then sent to the platform, which updates the global model based on the various model parameters and deploys the updated global model to the corresponding users to perform the analysis for the corresponding collaborative analysis needs. If deployed at the edge, the global model can also be lightweighted, such as through pruning, quantization, and knowledge distillation. The model parameters corresponding to the global model are then integrated into platform resource data.

[0042] A method for mining educational resource big data based on hybrid intelligence, the method includes: Based on user information, identify various potential educational analysis needs, filter these potential educational analysis needs, and obtain the target educational analysis needs. Acquire users' educational data information, conduct collaborative evaluation of various target educational analysis needs based on the educational data information, obtain the collaborative evaluation results of various target educational analysis needs, and mark the target educational analysis needs that require collaborative analysis according to the collaborative evaluation results as collaborative analysis needs.

[0043] Analyze the collaborative analysis needs of each user to obtain the corresponding collaborative data users and collaborative data range for each user; and perform collaborative processing based on the collaborative data users and collaborative data range for each user.

[0044] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0045] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A big data mining system for educational resources based on hybrid intelligence, characterized in that, Including both the platform side and the user side; The user terminal includes a resource analysis module; the platform terminal includes a resource information database, a demand analysis module, and a collaborative processing module. The resource analysis module is used to perform needs analysis on users, determine various potential educational analysis needs based on user information, filter each potential educational analysis need, and obtain each target educational analysis need. The system acquires users' educational data and conducts collaborative evaluations of various target educational analysis needs based on the educational data. It obtains the collaborative evaluation results for each target educational analysis need, which include those that do not require collaborative analysis and those that do. The system marks the target educational analysis needs that require collaborative analysis as collaborative analysis needs and sends the educational data and collaborative analysis needs to the platform. The resource information database is used to store educational data information of each user and platform resource data; The requirements analysis module analyzes the collaborative analysis requirements of each user to obtain the collaborative data users and the scope of collaborative data for each user. The collaborative processing module is used to perform collaborative processing based on the collaborative data users and the scope of collaborative data of each user, and to generate corresponding platform resource data according to the collaborative processing process, and send the platform resource data to the resource information database for storage.

2. The educational resource big data mining system based on hybrid intelligence according to claim 1, characterized in that, The communication connection between the platform and each user terminal.

3. The educational resource big data mining system based on hybrid intelligence according to claim 1, characterized in that, Based on user information, various potential educational analytics needs are identified, including: The platform establishes a reserve database, which is used to store various educational analysis needs and the applicable conditions corresponding to each educational analysis need. Establish a matching model, the expression of which is: ； In the formula: (s i ,YU) is the input data, s i This indicates the applicable conditions for the corresponding educational analysis needs in the reserve, where i represents the corresponding educational need, i = 1, 2, ..., n, and n is the number of educational analysis needs in the reserve; YU represents user information; s i →YU indicates that the applicable conditions match the user information; the output data is the matching value PU(s) i (YU), matching value is 1 or 0; The user information and the applicable conditions of the corresponding educational analysis needs in the reserve are integrated into the input data and input into the matching model for analysis to obtain the corresponding matching value; Each educational analytics requirement with a matching value of 1 is marked as a potential educational analytics requirement.

4. The educational resource big data mining system based on hybrid intelligence according to claim 1, characterized in that, Screening of various potential educational analytics needs, including: The system presents all potential educational analytics needs to users, who then filter and mark the remaining potential needs as target educational analytics needs.

5. The educational resource big data mining system based on hybrid intelligence according to claim 1, characterized in that, Screening of various potential educational analytics needs, including: Obtain the user's target requirement standards, collect features of each potential educational analysis need based on the target requirement standards, and obtain the target requirement features corresponding to each potential educational analysis need; calibrate the target requirement features of each potential educational analysis need through the target requirement standards, and mark each potential educational analysis need that passes the calibration as the target educational analysis need.

6. The educational resource big data mining system based on hybrid intelligence according to claim 1, characterized in that, Based on educational data, a collaborative assessment of the educational analysis needs for each objective is conducted, including: Identify the data requirements for each target educational analysis need, and determine whether the user's educational data meets the data requirements for each target educational analysis need based on the educational data information. Collaborative assessment results that do not meet the data requirements for the target education analysis are marked as requiring collaborative analysis; The platform updates the corresponding effectiveness standards for each educational analysis need in the reserve in real time; it estimates the expected effectiveness of the target educational analysis needs that meet the data requirements based on educational data information; it matches each effectiveness standard from the reserve according to the target educational analysis needs, and compares each effectiveness standard with the corresponding estimated effectiveness to obtain the effectiveness comparison results, which include the effectiveness being qualified and the effectiveness being unqualified. Mark the results of the collaborative assessment of the needs for analysis of the target education that do not meet the requirements as requiring collaborative analysis; Mark the results of the collaborative assessment of the target education needs analysis that meet the requirements as not requiring collaborative analysis.

7. The educational resource big data mining system based on hybrid intelligence according to claim 6, characterized in that, The comparison is made between each performance criterion and the corresponding predicted performance, including: Calculate the optimal value between the standard effect and the corresponding predicted effect, and label the obtained optimal value as YR. j j represents the corresponding effect standard, j = 1, 2, ..., m, where m is the number of effect standards; The comprehensive comparison value is calculated using the following formula: ； In the formula: UB is the comprehensive comparison value; b j This represents the corresponding proportionality coefficient, with a value range of (0, 1). When the overall comparison value is greater than the threshold X1, the evaluation effect is unqualified; The evaluation effect is qualified when the comprehensive comparison value is not greater than the threshold X1.

8. The educational resource big data mining system based on hybrid intelligence according to claim 3, characterized in that, The collaborative analytics needs of each user were analyzed, including: Obtain the corresponding effect criteria for each collaborative analysis requirement, and determine the requirement data information for the corresponding collaborative analysis requirement based on each effect criterion. Each collaborative analysis requirement and requirement data is input into the resource information database for matching and analysis to obtain candidate collaborative solutions that meet each collaborative analysis requirement. The candidate collaborative solutions are then filtered to obtain the target collaborative solution. Based on the target collaborative solution, the corresponding collaborative data users and the scope of collaborative data are determined.

9. A method for mining educational resource big data based on hybrid intelligence, characterized in that, The method applied to the hybrid intelligence-based big data mining system for educational resources as described in any one of claims 1 to 8 includes: Based on user information, identify various potential educational analysis needs, filter these potential educational analysis needs, and obtain the target educational analysis needs. Acquire users' educational data information, conduct collaborative evaluation of various target educational analysis needs based on the educational data information, obtain the collaborative evaluation results of various target educational analysis needs, and mark the target educational analysis needs that require collaborative analysis based on the collaborative evaluation results as collaborative analysis needs. Analyze the collaborative analysis needs of each user to obtain the corresponding collaborative data users and collaborative data range for each user; and perform collaborative processing based on the collaborative data users and collaborative data range for each user.

Images