Big data-based education business processing method and system

Abstract

The application relates to the technical field of intelligent education, and discloses an education business processing method and system based on big data, which comprises the following steps: standardizing and integrating original education data of an education management platform and a student learning terminal to obtain standardized education data; performing directional feature extraction on the standardized education data based on the business target of an education business to obtain a multi-dimensional business feature set of the education business; performing correlation analysis on the multi-dimensional business feature set to obtain a business index of the education business; performing business logic correlation between historical education business processing data and the business index to obtain a preliminary education business scheme of the education business; generating a multi-target optimization decision framework of the education business according to key constraint condition elements of the education business; and based on the multi-target optimization decision framework, iteratively improving the preliminary education business scheme to obtain a target education business scheme of the education business; and the application can improve the speed of education business processing based on big data.

Description

Technical Field

[0001] This invention relates to the field of intelligent education technology, and in particular to a method and system for processing education business based on big data. Background Technology

[0002] In the field of intelligent education technology, existing technologies lack a standardized integration process for raw educational data generated by education management platforms and student learning terminals. Raw data often suffers from formatting issues, redundancy, or missing key information, making it impossible to transform into usable standardized educational data. This undermines the reliable foundation for subsequent data-driven business processing, hinders the accurate matching of targeted feature extraction with educational business objectives, and results in biased construction of multi-dimensional business feature sets. Consequently, it becomes impossible to accurately extract key information from educational business, making it difficult to generate effective business indicators and impacting the overall accuracy of educational business processing.

[0003] Existing technologies have significant shortcomings in the generation and optimization of education business solutions. When linking historical education business processing data with business metrics, they lack a rigorous business logic association mechanism, making it difficult to fully leverage historical processing experience in the formation of initial education business solutions. Furthermore, existing technologies fail to consider key constraints of education business, cannot construct a targeted multi-objective optimization decision-making framework, and adjustments to the initial solution remain superficial, failing to achieve iterative improvement. The resulting business solutions are ultimately unable to meet the actual needs of education business, reducing processing efficiency and failing to guarantee the quality of results, thus making them ill-suited to the requirements of business development in intelligent education scenarios. Summary of the Invention

[0004] This invention provides a method and system for processing educational business based on big data, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a big data-based education business processing method, comprising: S1. Standardize and integrate the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational business. S2. Based on the business objectives of the education business, perform targeted feature extraction on the standardized education data to obtain a multi-dimensional business feature set for the education business; S3. Perform correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business; S4. Logically associate the historical education business processing data with the business indicators to obtain a preliminary education business plan; S5. Generate a multi-objective optimization decision framework for the education business based on the key constraints of the education business; S6. Based on the multi-objective optimization decision framework, the preliminary education business plan is iteratively improved to obtain the target education business plan.

[0006] In a preferred embodiment, the standardization and integration of the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational operations includes: Simultaneously collect raw educational data from the education management platform and student learning terminals; The original educational data is cleaned to obtain the cleaned data for the educational business. The cleaned data is standardized to obtain standardized format data for the education business. The standardized format data is fused to obtain the standardized educational data for the educational business.

[0007] In a preferred embodiment, the standardized educational data is subjected to targeted feature extraction based on the business objectives of the education business to obtain a multi-dimensional business feature set for the education business, including: The key business dimensions of the education business are analyzed to obtain the set of key dimensions of the education business. Based on the set of key dimensions, the data fields of the standardized education data are selected accordingly to obtain the original feature subset of the education business; The original feature subset is subjected to multidimensional structuring to obtain the multidimensional business feature set of the education business.

[0008] In a preferred embodiment, the step of performing correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business includes: By performing association matching on each feature dimension of the multi-dimensional business feature set, the association rule set of the education business is obtained; Based on the set of association rules, the degree of impact of the educational business on the target is evaluated to obtain the feature association strength of the educational business. The formula for calculating the feature association strength is as follows: ; In the formula, The feature association strength of the education business, For example, a single association rule in the set of association rules. For the association rule The historical significance of the impact on business objectives For the association rule Reliability under the current business context For the association rule The prevalence of the defined feature combination in the current business scenario is, in other words, the frequency of occurrence of the feature combination in the current business scenario. Based on the strength of the feature association, the multi-dimensional business feature set is filtered step by step to obtain the key feature subset of the education business; Business semantic mapping is performed on the subset of key features to obtain the business metrics of the education business.

[0009] In a preferred embodiment, the step of performing association matching on each feature dimension of the multi-dimensional business feature set to obtain the association rule set for the education business includes: Identify the potential business logic relationships of each feature dimension and generate the relationship analysis results of the education business; Based on the correlation analysis results, feature value matching is performed on the correlation of each feature dimension to obtain the feature combination pattern of the education business. By unifying the structure of the feature combination patterns, the set of association rules for the education business is obtained.

[0010] In a preferred embodiment, the step of logically associating historical education business processing data with the business indicators to obtain a preliminary education business plan includes: The historical education business processing data is mapped and categorized with the types of the business indicators to establish a correspondence between the historical education business processing data and the business indicators; Based on the correspondence, the business operation mode of the historical education business processing data is matched and filtered to obtain the historical business operation mode of the education business. Based on the historical business operation mode, a preliminary education business plan for the education business is generated.

[0011] In a preferred embodiment, generating a multi-objective optimization decision framework for the education business based on the key constraint elements of the education business includes: Based on the stated business objectives, determine the direction for optimizing the decision-making process of the education business; The key constraints of the education business are mapped and bound to the decision optimization direction to obtain the boundary constraints of the education business. By integrating the boundary constraints, a multi-objective optimization decision-making framework for the education business is obtained.

[0012] In a preferred embodiment, the step of iteratively refining the preliminary education business plan based on the multi-objective optimization decision framework to obtain the target education business plan includes: The preliminary education business plan is placed into the multi-objective optimization decision framework for performance deviation evaluation to obtain the evaluation results of the education business. Based on the evaluation results, the preliminary education business plan is optimized and adjusted to obtain the target education business plan.

[0013] In a preferred embodiment, optimizing and adjusting the preliminary education business plan based on the evaluation results to obtain the target education business plan includes: Root cause analysis is performed on the deviations in the evaluation results to obtain a deviation tracing report for the educational operations; Based on the deviation tracing report, the parameter configuration of the preliminary education business plan is reconstructed and corrected to obtain the reconstructed education business plan; The constraint compliance of the reconstruction scheme is verified to obtain the verification conclusion of the education business. When the verification conclusions meet all constraints, the verified reconstruction scheme will be determined as the target education business scheme for the education business.

[0014] To address the above problems, the present invention also provides an education business processing system based on big data, the system comprising: The data standardization and integration module is used to standardize and integrate the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational business. The targeted feature extraction module is used to extract targeted features from the standardized education data based on the business objectives of the education business, so as to obtain a multi-dimensional business feature set of the education business. The correlation analysis module is used to perform correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business. The business logic association module is used to associate historical education business processing data with the business indicators to obtain a preliminary education business plan. The decision framework construction module is used to generate a multi-objective optimization decision framework for the education business based on the key constraints of the education business. The solution iteration and optimization module is used to iteratively improve the preliminary education business solution based on the multi-objective optimization decision framework to obtain the target education business solution.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This technology ensures the integrity and consistency of standardized educational data by performing a standardized integration process on the raw educational data from the education management platform and student learning terminals, through synchronous collection, data cleaning, format standardization, and data fusion. This provides a high-quality data foundation for subsequent business processing. Simultaneously, based on the key dimension set analyzed from the educational business objectives, data fields are selected and subjected to multi-dimensional structured processing to form a precise multi-dimensional business feature set. Then, through association rule matching, feature association strength evaluation, and key feature screening, the key feature subset is mapped to business indicators, significantly improving the accuracy and relevance of business indicators and providing a reliable basis for generating educational business solutions.

[0016] 2. This technology maps and categorizes historical education business processing data with business indicators, establishing clear correspondences to match and filter suitable historical business operation patterns, ensuring the rationality of the initial education business plan. Based on this, it determines the direction of decision optimization by combining key constraints of education business and binds them to form boundary constraints, constructing a multi-objective optimization decision framework. Through performance deviation evaluation of the initial plan, root cause analysis of deviations, parameter configuration reconstruction, and constraint compliance verification, it iterative improvement is achieved. The final target education business plan accurately meets business needs, significantly improves the efficiency of education business processing, and ensures that the business processing results meet all constraints and business objectives, thereby improving the overall quality of education business processing. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an embodiment of the education business processing method based on big data provided by the present invention. Figure 2 A functional block diagram of an education business processing system based on big data provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a big data-based education business processing method. The executing entity of the big data-based education business processing method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application embodiment: a server, a terminal, etc. In other words, the big data-based education business processing method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a big data-based education business processing method according to an embodiment of the present invention. In this embodiment, the big data-based education business processing method includes: S1. Standardize and integrate the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational business. In this embodiment of the invention, the standardization and integration of the original educational data from the education management platform and student learning terminals to obtain standardized educational data for educational operations includes: Simultaneously collect raw educational data from the education management platform and student learning terminals; The original educational data is cleaned to obtain the cleaned data for the educational business. The cleaned data is standardized to obtain standardized format data for the education business. The standardized format data is fused to obtain the standardized educational data for the educational business.

[0021] Specifically, stable data connections are established with the education management platform and student learning terminals through pre-defined dedicated data interfaces. The scope of raw educational data to be collected is clearly defined, covering teaching arrangement information, teacher teaching progress records, and student attendance records from the education management platform, as well as homework completion and submission records, learning time statistics for each subject, and practice error records from the student learning terminals. Then, the same collection time interval is set, and data collection operations on both platforms and terminals are started simultaneously within this time interval. The stability of the data transmission link is monitored in real time. If a transmission interruption occurs, the collection request is immediately re-initiated to ensure that all raw educational data generated by the education management platform and student learning terminals within this time interval is completely obtained, ultimately yielding the raw educational data from the education management platform and student learning terminals.

[0022] Furthermore, the acquired raw educational data is examined one by one. First, missing fields are identified in each data entry. If the missing field is non-critical information, such as the learning notes field in the student learning terminal data, the data entry is retained, and the missing field status is marked. If the missing field is critical information, such as the student's unique identification information in the education management platform data, the data entry is directly removed from the raw dataset. Then, by comparing the unique identifier, student identification information, and data generation timestamp of each data entry, completely duplicate data entries are identified. Only one duplicate entry is retained, and the rest are deleted to remove redundant data.

[0023] Furthermore, based on the normal data range set for the education business, such as students' cumulative daily study time for each subject not exceeding a reasonable upper limit and the number of homework submissions being non-negative, abnormal data exceeding this range is filtered out. After manual verification to confirm that the data is indeed abnormal, it is removed, completing all data cleaning operations and obtaining the cleaned data for the education business.

[0024] Furthermore, a unified format specification for education business data is formulated, clearly defining the type of each data field, such as student identification information as text, learning duration as numeric, and homework submission time as date; the length requirements for each field are specified, with student identification information fixed to a specific character length; the format standards for each data are specified, converting numeric fields with learning duration units that do not meet the requirements to fixed units, and after processing, verifying that all fields of each data entry conform to the unified format specification, and obtaining standardized format data for education business after confirming that all conformity is met.

[0025] Furthermore, a unique student identification information was identified as the relevant field for data fusion. This field exists and is unique in both the standardized format data of the education management platform and the standardized format data of the student learning terminal. Using this field as the core, the standardized format data of the education management platform, such as student attendance records and teacher evaluations of student classroom performance, was matched with the standardized format data of the student learning terminal, such as student homework scores and access records of various learning modules. This merged the two types of data entries corresponding to the same student's unique identification information into a single comprehensive data entry containing both types of data information.

[0026] Furthermore, if there are cases where field names are the same but the content is inconsistent during the merging process, such as differences in student names recorded on the education management platform and student learning terminals, the corresponding field content in the standardized format data of the education management platform will be used for unified correction. After the merging is completed, each comprehensive data entry will be checked to ensure that it fully contains the key data information of the education management platform and student learning terminals, and that there are no logical conflicts between the data. After confirming that all requirements are met, standardized education data for education business will be obtained.

[0027] In general, data cleaning of raw education data to obtain cleaned data for education business requires verifying the fields of the original data, marking missing fields for non-critical information and retaining the data, and removing data for fields with missing critical information; then comparing the unique identifier information of the data to delete duplicate entries, and finally filtering out abnormal data according to the normal range of education business data, removing them after manual review and confirmation, thus completing the cleaning process to obtain cleaned data for education business.

[0028] In general, to standardize the format of cleaned data to obtain standardized format data for education business, it is necessary to first establish a unified format specification, clarifying the data field types, lengths, and format standards; then process each piece of cleaned data according to the specification, adjusting the date format, text encoding, and numerical units; after processing, verify the compliance of each field; and once confirmed to be error-free, obtain the standardized format data for education business.

[0029] In general, to obtain standardized educational data for educational business by fusing standardized format data, it is necessary to use the student's unique identification information as the associated field and match and merge the standardized format data of the education management platform and the student's learning terminal. If there is a conflict in the field content, the data of the education management platform shall prevail for correction. After merging, the data integrity and logical consistency shall be verified. Once the requirements are met, the standardized educational data for educational business is obtained.

[0030] S2. Based on the business objectives of the education business, perform targeted feature extraction on the standardized education data to obtain a multi-dimensional business feature set for the education business; In this embodiment of the invention, the step of extracting targeted features from the standardized educational data based on the business objectives of the educational business to obtain a multi-dimensional business feature set for the educational business includes: The key business dimensions of the education business are analyzed to obtain the set of key dimensions of the education business. Based on the set of key dimensions, the data fields of the standardized education data are selected accordingly to obtain the original feature subset of the education business; The original feature subset is subjected to multidimensional structuring to obtain the multidimensional business feature set of the education business.

[0031] Specifically, first, clarify the specific content of the business objectives for the education business. Starting from the logic and core needs for achieving these objectives, break down the key business dimensions that support their attainment. During this breakdown, analyze each core influencing factor required for objective achievement, ensuring that each decomposed dimension is directly related to the business objectives and covers the core aspects that need focus during the objective achievement process, excluding dimensions irrelevant to the business objectives. After the breakdown is complete, summarize all identified key business dimensions and manually verify that each dimension is a necessary support for achieving the business objectives, with no duplicates or redundancies, ultimately obtaining the set of key dimensions for the education business.

[0032] Furthermore, based on the established set of key dimensions for education operations, a thorough analysis is conducted on the data support required for each key dimension. This involves selecting specific data fields directly corresponding to each dimension from standardized education data. Each key dimension must be matched with at least one data field that accurately reflects its meaning, ensuring that all selected fields originate from standardized education data without omissions or incorrect selections. All selected fields corresponding to the key dimensions are then categorized and organized according to the key dimensions, forming field combinations based on each key dimension. It must be confirmed that each field combination fully supports the information presentation of its corresponding key dimension, ultimately yielding the original feature subset of the education operations.

[0033] Furthermore, for the original feature subset of the acquired education business, data grouping is performed according to the dimensions in the key dimension set. All data fields belonging to the same key dimension are grouped into the same data group, ensuring that each data group corresponds completely to a key dimension. Within each data group, the logical relationships between the data fields are analyzed, and the fields are arranged in a unified logical order to ensure clear logical coherence within the group.

[0034] Furthermore, a unique dimension identifier is added to each data group, clearly reflecting the name of the corresponding key dimension. An index is then created linking different data groups, based on the student's unique identification information. This index enables rapid querying of the same student's data across different data groups. After the complete processing of grouping, sorting, adding identifiers, and creating indexes, the original feature subset is transformed into a clearly structured, dimensionally defined, and logically connected set, ultimately yielding a multi-dimensional business feature set for education operations.

[0035] In summary, to obtain the key dimension set of education business by analyzing the key business dimensions of the business objectives, it is necessary to clarify the business objectives and core needs, break down the core impact levels that support the objectives from the perspective of implementation logic, eliminate irrelevant dimensions, summarize them, and manually review and confirm that there is no duplication or redundancy, so as to finally obtain the key dimension set of education business.

[0036] In summary, the original feature subset of education business is obtained by selecting corresponding data fields from standardized education data based on the key dimension set. It is necessary to analyze the data support requirements for each key dimension, filter corresponding fields in the standardized education data to ensure that each dimension matches valid fields, and then classify and organize the filtered fields by dimension to finally obtain the original feature subset of education business.

[0037] In summary, to obtain a multi-dimensional business feature set for education by performing multi-dimensional structuring on the original feature subset, the original feature subset needs to be classified and grouped according to key dimensions. Within each group, the field logic should be sorted and ordered. A unique identifier for each dimension should be added to each group. An inter-group association index should be established using the student's unique identification information to form a set with a clear structure, well-defined dimensions, and orderly data association, ultimately resulting in a multi-dimensional business feature set for education.

[0038] S3. Perform correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business; In this embodiment of the invention, the step of performing correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business includes: By performing association matching on each feature dimension of the multi-dimensional business feature set, the association rule set of the education business is obtained; Based on the set of association rules, the degree of impact of the educational business on the target is evaluated to obtain the feature association strength of the educational business. The formula for calculating the feature association strength is as follows: ; In the formula, The feature association strength of the education business, For example, a single association rule in the set of association rules. For the association rule The historical significance of the impact on business objectives For the association rule Reliability under the current business context For the association rule The prevalence of the defined feature combinations in the current business scenario, among which Iterate through all association rules in the association rule set.

[0039] Preferably, the degree of historical significance The determination method is as follows: Retrieve all data containing association rules from the historical education business database. Historical business processing records with combined features; for each historical record, calculate the contribution of the business processing result to the preset business objective. For example, if the business objective is to improve student grades, the contribution can be quantified as the percentage increase in grades; perform statistical analysis on the contribution of this rule in all historical records, for example, take the average value and normalize it, and use it as the rule's... In addition, decision analysis methods such as entropy weighting and analytic hierarchy process (AHP) can be used to assign values ​​based on the correlation between the rule and the final business results in historical data. value.

[0040] Based on the strength of the feature association, the multi-dimensional business feature set is filtered step by step to obtain the key feature subset of the education business; Business semantic mapping is performed on the subset of key features to obtain the business metrics of the education business.

[0041] The step of performing association matching on each feature dimension of the multi-dimensional business feature set to obtain the association rule set for the education business includes: Identify the potential business logic relationships of each feature dimension and generate the relationship analysis results of the education business; Based on the correlation analysis results, feature value matching is performed on the correlation of each feature dimension to obtain the feature combination pattern of the education business. By unifying the structure of the feature combination patterns, the set of association rules for the education business is obtained.

[0042] Specifically, we first analyze the business meaning of each feature dimension in the multi-dimensional business feature set, clarifying the specific information represented by each feature dimension in the education business scenario. Then, based on the core logic of the education business, we identify the potential business logic correlations between each feature dimension, such as the correlation between the learning duration features of different subjects and the homework completion quality features of the corresponding subjects. Next, based on the identified potential correlations, we match the feature values ​​of each feature dimension. For example, we correspond different numerical ranges of the learning duration feature with different levels of the homework completion quality feature to form feature combination patterns. Then, we unify the structure of all the formed feature combination patterns and organize all feature combination patterns using a fixed format of "feature dimension one feature dimension two - correlation description". Finally, we summarize to obtain the set of correlation rules for the education business.

[0043] Furthermore, we first analyze the historical application records corresponding to each association rule in the association rule set, and statistically analyze the impact of each rule on the achievement of education business goals in historical business processing. This determines the historical significance of each association rule's impact on business goals. Then, for the current actual scenario of education business, we statistically analyze the total number of times the feature combination described by each association rule appears in the current scenario and the number of times it conforms to the association relationship description. We calculate the proportion of the conformity count to the total count to determine the reliability of each association rule under the current business context. At the same time, we statistically analyze the frequency of occurrence of the feature combination defined by each association rule in the current business scenario to determine the universality of the feature combination. Then, we comprehensively analyze the historical significance, current reliability, and current universality of each association rule to evaluate the impact of each rule on business goals. Finally, we summarize the evaluation results of all association rules to form a quantitative judgment on the overall association of the multi-dimensional business feature set, and finally obtain the feature association strength of education business.

[0044] Furthermore, based on the objectives of the education business, a screening threshold for the strength of feature associations is first set. This threshold must ensure that the selected feature dimensions can effectively support the achievement of business objectives. Then, the feature association strength corresponding to each feature dimension in the multi-dimensional business feature set is compared with the screening threshold. Feature dimensions with association strength higher than the threshold are retained, completing the first level of screening. Next, the interrelationships between the feature dimensions retained after the first level of screening are analyzed, and the association strength of these feature dimensions after pairwise combinations is calculated. Feature dimensions whose combined association strength is lower than the individual association strength and has no benefit to the overall association effect are eliminated, completing the second level of screening. The above screening process is repeated. Each level of screening analyzes the association strength and combination effect of feature dimensions based on the results of the previous level of screening, until the association strength of the retained feature dimensions after screening can stably meet the business objective requirements and there are no redundant feature dimensions. Finally, the key feature subset of the education business is obtained.

[0045] Furthermore, we first analyze the technical description of each feature dimension in the key feature subset, transforming the technical feature definitions into easily understandable business semantics in the context of education business. For example, we transform "the average daily data transmission volume corresponding to the subject code" into "the average daily access volume of learning resources for the corresponding subject." Then, we match the transformed business semantics with the core objectives of education business, clarifying the role of each business semantic in measuring the achievement of business objectives. For example, "the average daily access volume of learning resources for the corresponding subject" corresponds to the business objective item of measuring "learning participation in the corresponding subject." Next, we standardize the naming of these business semantics, forming a standard expression using the format of "business objective item - measurement dimension." For example, we standardize "the average daily access volume of learning resources for the corresponding subject" as "learning participation in the corresponding subject - average daily resource access volume." Finally, we summarize all the standardized expressions to form indicators that can be directly used for education business evaluation and decision-making, ultimately obtaining the business indicators for education business.

[0046] Specifically, This refers to the feature correlation strength of education business, which is used to comprehensively reflect the degree of comprehensive correlation between each correlation rule in the multi-dimensional business feature set and the education business objectives. It is the result obtained by integrating and calculating the relevant data of the correlation rules, and provides a basis for subsequent step-by-step screening of the multi-dimensional business feature set. Refers to association rules The historical significance of the impact on business objectives stems from the association rules. In the process of handling historical education business, the application of record sorting and analysis, by statistically analyzing the specific performance of this association rule in promoting the achievement of education business goals or influencing the effectiveness of business goal achievement, determines the historical significance of its impact on business goals. Refers to association rules The reliability of the validity under the current business context is determined by statistical association rules in the current education business scenario. The total number of occurrences of the described feature combinations, and the number of times they conform to the association relationship defined by the association rule, are used to determine the proportion obtained by dividing the number of times they conform to the association relationship by the total number of occurrences. This refers to the sum of squares of the validity reliability of all association rules. It is obtained by collecting the validity reliability data of each association rule, squaring the validity reliability data of each association rule, and then summing all the squared results. Refers to association rules The prevalence of the defined feature combinations in the current business scenario is determined by statistical association rules in the current education business scenario. The frequency of the defined feature combination is used to determine its prevalence in the current scenario.

[0047] Specifically, we first analyze the business meaning of each feature dimension in the multi-dimensional business feature set, clarifying the specific information each feature dimension represents in the education business scenario. For example, one feature dimension represents the student's subject learning time, while another feature dimension represents the accuracy rate of homework completion for that subject. Then, combined with the core operational logic of the education business, such as the intrinsic relationship between learning input and learning output, and the mutual influence between classroom participation and knowledge mastery, we determine whether there are direct or indirect business relationships between the feature dimensions. We record the determined relationships in detail, including the names of the associated feature dimensions and the business logic on which the relationships are based, and finally generate the correlation analysis results of the education business.

[0048] Furthermore, based on the correlation analysis results of the generated education business, the characteristic value range of each involved feature dimension is first determined. For example, the characteristic value of the feature dimension representing learning time is divided into different time intervals, and the characteristic value of the feature dimension representing homework completion accuracy is divided into different percentage levels. Then, according to the business correlation logic clearly defined in the correlation analysis results, the characteristic values ​​of the corresponding feature dimensions are matched. For example, when the characteristic value of the learning time feature dimension is in a specific time interval, the characteristic value of the homework completion accuracy feature dimension is in a specific percentage level. This correspondence of characteristic values ​​is recorded completely to form the feature combination pattern of the education business.

[0049] Furthermore, a unified structural format for feature combination patterns is first established. This format must include the names of the feature dimensions involved in the association, the feature value ranges of each feature dimension, and a description of the association relationships between the feature dimensions. The expression methods for each part are standardized; for example, feature dimension names should use full names, feature value ranges should use the unified format "XX-XX," and the descriptions of association relationships should use concise and clear declarative sentences. Then, according to this unified format, all existing feature combination patterns for education business are adjusted, correcting any non-standard expressions or inconsistent structures to ensure complete uniformity in format and expression. Finally, all adjusted feature combination patterns are summarized and organized to form a consistent and clearly defined set, resulting in the set of association rules for education business.

[0050] In summary, to obtain the association rule set for education business by performing correlation matching on each feature dimension of the multi-dimensional business feature set, it is necessary to first identify the potential business logic correlation between each feature dimension and generate the correlation analysis results of education business, then perform feature value matching on the correlation of each feature dimension based on the analysis results to obtain the feature combination pattern of education business, and finally unify the structure of the feature combination pattern to form the final association rule set for education business.

[0051] In summary, to assess the impact of a set of association rules on the goals of education business and obtain the feature association strength of education business, it is necessary to combine the historical significance of each rule's impact on the business goals, the reliability of its validity under the current business context, and the prevalence of the feature combinations defined by the rules in the current business scenario. By comprehensively evaluating and summarizing these factors, the target impact of each rule can be obtained, and finally the feature association strength of education business can be obtained.

[0052] In summary, to obtain the key feature subset of education business by progressively filtering the multi-dimensional business feature set based on the strength of feature association, it is necessary to first set a feature association strength filtering threshold that meets the requirements of education business objectives, compare the association strength of each feature dimension with the threshold to complete the initial filtering, then analyze the combined association effect between the filtered feature dimensions and eliminate redundant dimensions. After progressive filtering, until the retained feature dimensions can stably support the business objective requirements, the key feature subset of education business is finally obtained.

[0053] In summary, to obtain business metrics for education business by mapping key feature subsets to business semantics, it is necessary to transform the technical descriptions of each feature dimension in the key feature subset into business semantics in the context of education business, clarify the correspondence between each business semantic and business objective and standardize the naming, and finally summarize all the standardized business semantic expressions to form the business metrics for education business.

[0054] In summary, to identify the potential business logic relationships among various feature dimensions and generate correlation analysis results for education business, it is necessary to first analyze the business meaning of each feature dimension in the multi-dimensional business feature set, clarify the specific information represented by each feature dimension in the education business scenario, and then, in combination with the core operating logic of education business, determine whether there are direct or indirect business relationships between each feature dimension. The determined relationships should be recorded in detail, and finally, correlation analysis results for education business should be generated.

[0055] In summary, to obtain the feature combination pattern of education business by matching the feature values ​​of each feature dimension based on the correlation analysis results, it is necessary to first determine the feature value range of each feature dimension involved in the correlation, and then match the feature values ​​of the corresponding feature dimensions according to the business correlation logic clearly defined in the correlation analysis results. The correspondence between the feature values ​​of different feature dimensions is recorded, such as the correspondence between a specific feature value range of a feature dimension and a specific feature value level of another related feature dimension. Finally, the feature combination pattern of education business is obtained.

[0056] In summary, to unify the structure of feature combination patterns to obtain the set of association rules for education business, it is necessary to first establish a unified structural format for feature combination patterns. This format should standardize the description of the feature dimension names involved in the association, the presentation of the feature value range of each feature dimension, and the description of the relationship between feature dimensions. Then, all the obtained feature combination patterns for education business should be adjusted according to this unified format, and content that is not standardized or inconsistent in structure should be corrected. Finally, the adjusted feature combination patterns should be summarized and organized to form a set with a consistent structure and clear content, which ultimately yields the set of association rules for education business.

[0057] S4. Logically associate the historical education business processing data with the business indicators to obtain a preliminary education business plan for the education business. In this embodiment of the invention, the step of logically associating historical education business processing data with the business indicators to obtain a preliminary education business plan includes: The historical education business processing data is mapped and categorized with the types of the business indicators to establish a correspondence between the historical education business processing data and the business indicators; Based on the correspondence, the business operation mode of the historical education business processing data is matched and filtered to obtain the historical business operation mode of the education business. Based on the historical business operation mode, a preliminary education business plan for the education business is generated.

[0058] Specifically, we first need to clarify the specific types of data processed in history education, including homework correction records for each class in a semester, student participation data in each subject over the past three years, frequency and feedback data on the use of past teaching resources, and data on the implementation process of history teaching and tutoring programs. At the same time, we need to determine the specific types of business indicators, including student subject knowledge mastery achievement rate indicators, student learning participation indicators, teaching resource utilization efficiency indicators, and teaching and tutoring effectiveness indicators.

[0059] Furthermore, based on the correlation logic between the information contained in various types of historical education business processing data and business indicators, historical data is mapped and categorized with business indicators. For example, in the homework correction records of each class in a certain semester, information such as the accuracy rate of students' homework and the distribution of error types can directly reflect the students' mastery of mathematical knowledge. Therefore, it is mapped with the student subject knowledge mastery achievement rate indicator. In the past teaching resource usage frequency and effect feedback data, information such as the number of times resources are used and student feedback scores can reflect the resource utilization effect. Therefore, it is mapped with the teaching resource utilization efficiency indicator. All historical data types and business indicator types are matched one by one to ensure that each historical data type can correspond to the relevant business indicator type. Finally, the correspondence between historical education business processing data and business indicators is established.

[0060] Furthermore, from the established correspondence, all historical education business processing data corresponding to each business indicator type are extracted. For these historical data, the business operation processes, specific operation methods, and key execution parameters contained therein are sorted out. For example, from the historical learning participation data corresponding to the student learning participation indicator, the operation process of "organizing students to sign in online during the morning reading time every day, pushing audio of subject knowledge points of fixed duration after signing in, and calculating the sign-in rate every week and reminding students who do not meet the standard" is sorted out. The operation method includes "using the online sign-in platform and selecting the channel for pushing knowledge point audio", and the execution parameters include "fixed time period for morning reading sign-in, fixed duration of audio, and reminders at a fixed time every week".

[0061] Furthermore, based on the core objectives of the current education business, such as "improving the participation of first-year junior high school students in Chinese language learning this semester", the identified business operation models are matched and screened. Those operation models that "can increase the participation of first-year junior high school students in Chinese language learning by a certain percentage" in historical applications are retained, while invalid operation models that "increase the attendance rate but do not increase the actual learning time of students" are eliminated. After screening, the historical business operation models of the education business are summarized.

[0062] Furthermore, the historical business operation model was broken down into its core components. For example, the historical model of "online check-in - audio push - check-in rate statistics - reminders for students who did not meet the standards" was broken down into four core components: "check-in rule setting, knowledge point audio production and push, check-in data statistics, and reminders for students who did not meet the standards." Considering the current educational business scenario, such as the fact that first-year junior high school Chinese classes primarily use tablets, the "knowledge point audio production and push" component was adjusted to "producing high-definition audio adapted for tablet playback and pushing it through the class learning group"; and since schools currently require reduced telephone communication, the "reminders for students who did not meet the standards" component was adjusted to "sending personalized reminder messages through the class learning platform, while also copying parents."

[0063] Furthermore, the adjusted core processes are integrated according to business logic, clarifying the executing entity for each process. For example, "the sign-in rule setting is the responsibility of the Chinese teacher, audio production is assisted by the teaching resources group, sign-in data statistics are performed by the homeroom teacher, and reminder messages are sent by the Chinese teacher." Execution time nodes are determined, such as "sign-in rules are updated at a fixed time every week, audio is pushed out before a fixed time on a fixed weekday morning, sign-in data is statistically analyzed at a fixed time every day morning, and reminder messages are sent at a fixed time every day morning." Execution requirements are clarified, such as "audio content must be aligned with the day's teaching progress, and reminder messages must include the dates students did not sign in and suggested improvement methods." This forms a complete operational plan adapted to the current scenario, ultimately resulting in a preliminary education business plan.

[0064] In summary, to map and categorize historical education business processing data with business indicator types to establish a corresponding relationship, it is necessary to first clarify the types of historical data and business indicator types, then match them one by one according to the business logic relationship between the two, ensuring that each historical data type corresponds to a relevant business indicator type, and finally establish the correspondence between the two.

[0065] In summary, to filter historical business operation patterns based on correspondence, it is necessary to extract historical data corresponding to each business indicator type, sort out the business operation process and key parameters, combine the current core objectives of education business, retain the operation patterns that can support the achievement of indicators, eliminate inconsistent patterns, and summarize them to obtain the historical business operation patterns of education business.

[0066] In general, to generate a preliminary education business plan based on historical business operation patterns, it is necessary to first break down its core links and clarify key contents, adjust the operation methods according to the current business scenario, integrate the links according to business logic, clarify the execution entity, time nodes and requirements, form a complete operation plan adapted to the scenario, and finally obtain the preliminary education business plan.

[0067] S5. Generate a multi-objective optimization decision framework for the education business based on the key constraints of the education business. In this embodiment of the invention, generating a multi-objective optimization decision framework for the education business based on the key constraint elements of the education business includes: Based on the stated business objectives, determine the direction for optimizing the decision-making process of the education business; The key constraints of the education business are mapped and bound to the decision optimization direction to obtain the boundary constraints of the education business. By integrating the boundary constraints, a multi-objective optimization decision-making framework for the education business is obtained.

[0068] Specifically, first clarify the specific content of the business objectives of the education business. For example, if the business objective is "to improve the average score of students in a certain grade in a certain subject", then break down the core directions that need to be optimized to achieve the objective. These directions must directly serve the achievement of the objective. For example, around "improving subject scores", we can break down the directions into "optimizing the allocation of teaching time for this subject in the daily courses", "adjusting the priority of the allocation of tutoring resources for this subject", and "improving the design of the question types and difficulty of homework for this subject". Each of the decomposed directions must have a direct causal relationship with the business objective, and directions that are irrelevant to the objective must be excluded. Through manual review, confirm that all directions can support the achievement of the business objective, and finally determine the decision optimization directions for the education business.

[0069] Furthermore, first, identify the key constraints of the education business. These constraints include insurmountable limitations during the operation of the education business, such as "the total daily teaching time does not exceed the prescribed limit," "the number of full-time teachers for this subject is fixed," "the total amount of available tutoring materials for this subject is limited," and "the total daily homework time for students does not exceed the prescribed standard." Each constraint must have a clearly defined scope and requirement. Then, for each identified optimization direction, match the corresponding key constraints one by one. For example, the direction of "optimizing the allocation ratio of teaching time for this subject" corresponds to the constraint of "the total daily teaching time does not exceed the prescribed limit," and the direction of "adjusting the priority of tutoring resource allocation for this subject" corresponds to the constraint of "the total amount of available tutoring materials for this subject is limited." Clarify the constraints that each optimization direction must follow. Record the correspondence between directions and elements and the constraints in a fixed format, such as "Optimization direction - optimize the allocation ratio of teaching time for this subject: the total daily teaching time must not exceed the prescribed limit." This ultimately yields the boundary constraints of the education business.

[0070] Furthermore, all obtained boundary constraints are categorized and organized according to the direction of decision optimization. For example, boundary constraints belonging to the direction of "optimizing the allocation ratio of teaching time for this subject" are grouped into one category, and boundary constraints belonging to the direction of "adjusting the priority of tutoring resource allocation for this subject" are grouped into another category, ensuring that each direction of decision optimization has a corresponding boundary constraint category. Within each category, the logical relationships between boundary constraints are clarified. For example, under the direction of "optimizing the allocation ratio of teaching time for this subject," the constraints of "the total daily teaching time does not exceed the prescribed upper limit" and "the duration of students' attention in class for this subject is limited" need to be considered in a coordinated manner to avoid conflicts.

[0071] Furthermore, the relationships between boundary constraints corresponding to different decision optimization directions are analyzed across categories. For example, the constraint of "limited total amount of tutoring materials" in the direction of "adjusting the priority of tutoring resource allocation for this subject" will affect the planning of material usage in the direction of "improving the design of homework question types and difficulty for this subject". All categories and relationships are integrated in a structured form, clarifying the scope of boundary constraints for each decision optimization direction and the mutual influence between constraints, forming a complete, conflict-free and logically clear framework, and finally obtaining a multi-objective optimization decision framework for education business.

[0072] In general, determining the direction of decision optimization based on the business objectives of education requires first clarifying the specific content of the business objectives, starting from the logic of achieving the objectives, breaking down the core optimization directions that can directly support the achievement of the objectives, excluding directions that are irrelevant to the objectives, and confirming through manual review that all decomposed directions are necessary support for achieving the objectives and are not redundant, and finally determining the direction of decision optimization for education business.

[0073] In general, mapping and binding the key constraints of education operations with decision optimization directions to obtain boundary constraints requires first identifying the key constraints that cannot be broken in the course of business and clarifying their limitations and requirements. Then, for each decision optimization direction, the corresponding key constraints are matched one by one, clarifying the constraints that each direction must follow. The correspondence between directions and elements and the content of constraints are recorded in a fixed format, ultimately yielding the boundary constraints of education operations.

[0074] In summary, to integrate boundary constraints into a multi-objective optimization decision-making framework, it is necessary to first classify and organize all boundary constraints according to the direction of decision optimization, sort out the logical relationships between constraints within each category, and then sort out the relationships between constraints in different directions across categories to avoid constraint conflicts. All categories and relationships are then integrated in a structured form to clarify the scope of constraints in each direction and the mutual influence between constraints, forming a complete and logically clear framework, ultimately resulting in a multi-objective optimization decision-making framework for education business.

[0075] S6. Based on the multi-objective optimization decision framework, the preliminary education business plan is iteratively improved to obtain the target education business plan.

[0076] In this embodiment of the invention, the step of iteratively improving the preliminary education business plan based on the multi-objective optimization decision framework to obtain the target education business plan includes: The preliminary education business plan is placed into the multi-objective optimization decision framework for performance deviation evaluation to obtain the evaluation results of the education business. Based on the evaluation results, the preliminary education business plan is optimized and adjusted to obtain the target education business plan.

[0077] The step of optimizing and adjusting the preliminary education business plan based on the evaluation results to obtain the target education business plan includes: Root cause analysis is performed on the deviations in the evaluation results to obtain a deviation tracing report for the educational operations; Based on the deviation tracing report, the parameter configuration of the preliminary education business plan is reconstructed and corrected to obtain the reconstructed education business plan; The constraint compliance of the reconstruction scheme is verified to obtain the verification conclusion of the education business. When the verification conclusions meet all constraints, the verified reconstruction scheme will be determined as the target education business scheme for the education business.

[0078] Specifically, the multi-objective optimization decision-making framework clarifies all boundary constraints and optimization directions. Boundary constraints include daily total teaching time limits, total teaching resource limits, and maximum student learning time limits per session. Optimization directions include improving student knowledge mastery rates, increasing the efficiency of teaching resource utilization, and enhancing student learning participation. The preliminary educational business plan is broken down into specific execution steps and parameter configurations. Execution steps include subject teaching arrangements, tutoring resource allocation, and homework assignment and grading. Parameter configurations cover the duration, frequency, and resource quantity of each step.

[0079] Furthermore, by comparing the plan with the boundary constraints in the framework, each step and parameter of the plan is checked to see if it meets the constraints. This includes determining whether the total teaching time for each subject exceeds the daily total teaching time limit and whether the amount of tutoring resources allocated is within the total resource limit. Simultaneously, by comparing the plan with the optimization direction, each step is checked to see if it supports the optimization direction and whether the types and difficulty of homework assignments help improve students' knowledge mastery achievement rate. All steps and parameters in the plan that do not meet the constraints or support the optimization direction are recorded, noting the specific manifestations and degrees of deviation, such as "the teaching time for a certain subject exceeds the daily total time limit by a certain amount" or "the homework question type is too simple, resulting in an inability to effectively improve the knowledge mastery achievement rate." A complete evaluation record document is then created, ultimately yielding the evaluation results for the educational business.

[0080] Furthermore, all deviations are extracted from the evaluation results of the education business, and the specific links and parameters in the preliminary education business plan corresponding to each deviation are clarified. For example, the deviation "the teaching time of a certain subject exceeds the daily total time limit" corresponds to the "daily teaching time parameter of the subject" in the plan, and the deviation "the monotony of homework question types leads to the inability to effectively improve the knowledge mastery pass rate" corresponds to the "homework question type configuration link" in the plan.

[0081] Furthermore, for each deviation item, an adjustment plan is formulated by combining the boundary constraints and decision optimization direction of the multi-objective optimization decision framework. If the duration exceeds the constraint, the complexity of the teaching content of this subject and other subjects and the students' acceptance ability are recalculated, and the daily teaching time of this subject is reduced by a certain amount. The reduced time is then reasonably allocated to subjects with more complex teaching content. If the homework question type is too simple, the question types that can effectively improve the knowledge mastery pass rate are referenced, and application questions, extension questions, and other question types are added, and the proportion of different question types is adjusted.

[0082] Furthermore, all the adjusted steps and parameters are integrated into a new solution. A comprehensive check is then conducted against the framework's boundary constraints and decision optimization directions to confirm that no deviation items are missing and that all adjustments meet the requirements. This ensures that the new solution can support the realization of all decision optimization directions without breaking any boundary constraints, ultimately yielding the target education business solution.

[0083] Specifically, first, review all deviations recorded in the evaluation results, clarifying the specific manifestations of each deviation. For example, in the preliminary education business plan, "the participation rate in after-school tutoring for a certain subject is lower than expected" or "the completion time for homework correction for a certain grade exceeds the standard." For each deviation, conduct root cause investigation from three levels: business logic design, resource allocation, and execution. At the business logic design level, check whether there is a logical disconnect between the corresponding plan steps and the business objectives. For example, is the low tutoring participation rate due to a conflict between tutoring time and students' after-school schedules? At the resource allocation level, confirm whether there is insufficient resource supply or mismatch of resource types. For example, is the homework correction time overdue due to an insufficient number of teachers? At the execution level, verify whether there are any operational deficiencies or information transmission deviations during the implementation of the plan. For example, are students not effectively notified of tutoring times?

[0084] Furthermore, the investigation process for each deviation, the identified root causes, and the corresponding evidence are recorded and organized into a well-structured document, ultimately resulting in a deviation tracing report for educational operations.

[0085] Furthermore, the root cause of each deviation item is extracted from the deviation source tracing report, such as "the root cause of low tutoring participation rate is the conflict between tutoring time and students' after-school schedule" and "the root cause of homework correction timeout is the insufficient number of correction teachers".

[0086] Furthermore, for each root cause, the relevant parameter configurations in the initial education business plan were adjusted accordingly: for the tutoring time conflict issue, a new survey was conducted on students' free time outside of school, and the tutoring time for that subject in the plan was adjusted to be during free time; for the problem of insufficient number of teachers to grade assignments, the total amount of homework for that grade and the teachers' grading efficiency were recalculated, and the number of suitable grading teachers was increased or the homework grading division mode was adjusted. During the adjustment process, it was ensured that the modification of each parameter configuration directly solved the corresponding root cause, and that there were no logical conflicts between the modified parameters. All the adjusted parameters were integrated into a new plan document, ultimately resulting in a restructured education business plan.

[0087] Furthermore, all key constraints on the education operations must be clearly defined, including "the total daily tutoring time for all subjects does not exceed the prescribed upper limit," "the average amount of homework graded per teacher does not exceed a reasonable threshold," and "the duration of a single tutoring session for students does not exceed a fatigue threshold." Each constraint must have a specific limiting standard. Against these constraints, the parameter configurations in the restructuring plan are verified one by one: check whether the total tutoring time for all subjects in the restructuring plan meets the upper limit requirement, verify whether the adjusted amount of homework graded by teachers is within a reasonable threshold, and confirm whether the duration of a single tutoring session for students does not exceed the fatigue threshold. The verification process and results for each constraint are recorded in detail. If all parameter configurations meet the constraints, record "the restructuring plan satisfies all constraints"; if some parameters do not meet the constraints, record the specific non-compliant constraints and their corresponding parameters, ultimately forming the verification conclusion for the education operations.

[0088] Further, examine the specific content of the verification conclusion to confirm whether it explicitly states that "the refactoring solution meets all constraints." If this is explicitly stated in the verification conclusion, conduct a final review of the refactoring solution to reconfirm that all parameter configurations in the solution do not deviate from the constraints, and that the overall logic of the solution is coherent, implementable, and free of unresolved deviations. After verification, the verified refactoring solution is officially adopted as the target educational business solution, ensuring that the solution effectively supports the achievement of educational business objectives and that no new deviations arise due to constraint issues during subsequent execution.

[0089] In summary, to evaluate the performance deviations of a preliminary education business plan within a multi-objective optimization decision-making framework and obtain the evaluation results, it is necessary to first clarify all the boundary constraints and decision optimization directions included in the framework. Then, the preliminary plan should be broken down into specific execution steps and parameter configurations. Each step and parameter of the plan should be checked against the boundary constraints in the framework to see if it meets the requirements. At the same time, the plan steps should be checked against the decision optimization directions to see if they can support the realization of the optimization directions. Steps and parameters in the plan that do not meet the constraints or do not support the optimization directions should be recorded, and the specific manifestations and degrees of deviations should be marked to form a complete evaluation record document. Finally, the evaluation results of the education business can be obtained.

[0090] In summary, optimizing and adjusting the preliminary education business plan based on the evaluation results to obtain the target education business plan requires extracting all deviations from the evaluation results, clarifying the specific steps and parameters in the preliminary plan corresponding to each deviation, and developing adjustment plans for each deviation based on the framework's boundary constraints and decision optimization directions. The adjusted steps and parameters are then integrated to form a new plan, which is then comprehensively checked against the framework to confirm that there are no deviations or omissions and that all adjustments meet the requirements. This ensures that the new plan can support the realization of all decision optimization directions without breaking any boundary constraints, ultimately yielding the target education business plan.

[0091] In general, to conduct root cause analysis on deviations in the evaluation results to obtain a deviation tracing report for education operations, it is necessary to first sort out the specific manifestations of all deviations in the evaluation results, and then investigate the root causes of each deviation from three levels: business logic design, resource allocation, and execution. For example, check whether the deviation is caused by a disconnect between business logic and objectives, insufficient resource supply, or inadequate execution. The investigation process, the identified root causes, and the corresponding evidence should be recorded in detail to form a well-structured document, and finally, a deviation tracing report for education operations will be obtained.

[0092] In summary, the parameter configuration of the preliminary education business plan is restructured and corrected based on the deviation source report to obtain the education business restructuring plan. It is necessary to extract the root cause corresponding to each deviation item from the deviation source report, adjust the relevant parameter configuration in the preliminary plan for each root cause, ensure that the parameter modification can directly solve the corresponding root cause, and that there is no logical conflict between the adjusted parameters. All the adjusted parameters are integrated into a new plan document, and finally the education business restructuring plan is obtained.

[0093] In summary, to obtain the verification conclusion for the education business by verifying the constraint compliance of the restructuring plan, it is necessary to first identify all key constraint elements and specific limiting standards of the education business, then verify the parameter configuration in the restructuring plan one by one against these constraints, check whether the parameters meet the constraint requirements, and record the verification process and results of each constraint in detail. If all parameters meet the constraints, record the conclusion that the constraints are met; if there are any non-compliance items, record the specific non-compliance items. Finally, the verification conclusion for the education business is obtained.

[0094] In general, when the verification conclusions meet all constraints, the verified refactoring scheme is determined as the target education business scheme for the education business. It is necessary to first confirm that the verification conclusions clearly record that the refactoring scheme meets all constraints, and then conduct a final review of the refactoring scheme to confirm that the scheme parameters do not deviate from the constraints and that the overall logic is coherent and feasible. After the review is correct, the verified refactoring scheme is officially determined as the target education business scheme for the education business.

[0095] like Figure 2 The diagram shown is a functional block diagram of an education business processing system based on big data provided in an embodiment of the present invention.

[0096] The big data-based education business processing system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the big data-based education business processing system 100 may include a data normalization and integration module 101, a targeted feature extraction module 102, a correlation analysis module 103, a business logic correlation module 104, a decision framework construction module 105, and a scheme iteration and optimization module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0097] In this embodiment, the functions of each module / unit are as follows: The data standardization and integration module 101 is used to standardize and integrate the original educational data from the education management platform and student learning terminals to obtain standardized educational data for educational business. The targeted feature extraction module 102 is used to extract targeted features from the standardized education data based on the business objectives of the education business, so as to obtain a multi-dimensional business feature set of the education business. The correlation analysis module 103 is used to perform correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business. The business logic association module 104 is used to associate historical education business processing data with the business indicators in a business logic manner to obtain a preliminary education business plan for the education business. The decision framework construction module 105 is used to generate a multi-objective optimization decision framework for the education business based on the key constraint elements of the education business. The iterative optimization module 106 is used to iteratively improve the preliminary education business plan based on the multi-objective optimization decision framework to obtain the target education business plan. In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for example, the division of modules is only a logical functional division, and other division methods may exist in actual implementation.

[0098] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0099] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0100] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0101] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions 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 solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A big data-based method for processing educational business, characterized in that: The method includes: S1. Standardize and integrate the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational business. S2. Based on the business objectives of the education business, perform targeted feature extraction on the standardized education data to obtain a multi-dimensional business feature set for the education business; S3. Perform correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business; S4. Logically associate the historical education business processing data with the business indicators to obtain a preliminary education business plan; S5. Generate a multi-objective optimization decision framework for the education business based on the key constraints of the education business; S6. Based on the multi-objective optimization decision framework, the preliminary education business plan is iteratively improved to obtain the target education business plan.

2. The education business processing method based on big data as described in claim 1, characterized in that, The process of standardizing and integrating the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational operations includes: Simultaneously collect raw educational data from the education management platform and student learning terminals; The original educational data is cleaned to obtain the cleaned data for the educational business. The cleaned data is standardized to obtain standardized format data for the education business. The standardized format data is fused to obtain the standardized educational data for the educational business.

3. The education business processing method based on big data as described in claim 1, characterized in that, Based on the business objectives of the education business, targeted feature extraction is performed on the standardized education data to obtain a multi-dimensional business feature set for the education business, including: The key business dimensions of the education business are analyzed to obtain the set of key dimensions of the education business. Based on the set of key dimensions, the data fields of the standardized education data are selected accordingly to obtain the original feature subset of the education business; The original feature subset is subjected to multidimensional structuring to obtain the multidimensional business feature set of the education business.

4. The education business processing method based on big data as described in claim 1, characterized in that, The correlation analysis of the multi-dimensional business feature set to obtain the business indicators of the education business includes: By performing association matching on each feature dimension of the multi-dimensional business feature set, the association rule set of the education business is obtained; Based on the set of association rules, the degree of impact of the educational business on the target is evaluated to obtain the feature association strength of the educational business. The formula for calculating the feature association strength is as follows: ； In the formula, The feature association strength of the education business, For example, a set of association rules. For the association rule The historical significance of the impact on business objectives For the association rule Reliability under the current business context For the association rule The prevalence of the defined feature combinations in the current business scenario; Based on the strength of the feature association, the multi-dimensional business feature set is filtered step by step to obtain the key feature subset of the education business; Business semantic mapping is performed on the subset of key features to obtain the business metrics of the education business.

5. The education business processing method based on big data as described in claim 4, characterized in that, The step of performing association matching on each feature dimension of the multi-dimensional business feature set to obtain the association rule set for the education business includes: Identify the potential business logic relationships of each feature dimension and generate the relationship analysis results of the education business; Based on the correlation analysis results, feature value matching is performed on the correlation of each feature dimension to obtain the feature combination pattern of the education business. By unifying the structure of the feature combination patterns, the set of association rules for the education business is obtained.

6. The education business processing method based on big data as described in claim 1, characterized in that, The step of logically associating historical education business processing data with the business indicators to obtain a preliminary education business plan includes: The historical education business processing data is mapped and categorized with the types of the business indicators to establish a correspondence between the historical education business processing data and the business indicators; Based on the correspondence, the business operation mode of the historical education business processing data is matched and filtered to obtain the historical business operation mode of the education business. Based on the historical business operation mode, a preliminary education business plan for the education business is generated.

7. The education business processing method based on big data as described in claim 1, characterized in that, The step of generating a multi-objective optimization decision framework for the education business based on the key constraints of the education business includes: Based on the stated business objectives, determine the direction for optimizing the decision-making process of the education business; The key constraints of the education business are mapped and bound to the decision optimization direction to obtain the boundary constraints of the education business. By integrating the boundary constraints, a multi-objective optimization decision-making framework for the education business is obtained.

8. The education business processing method based on big data as described in claim 1, characterized in that, The process of iteratively refining the preliminary education business plan based on the multi-objective optimization decision framework to obtain the target education business plan includes: The preliminary education business plan is placed into the multi-objective optimization decision framework for performance deviation evaluation to obtain the evaluation results of the education business. Based on the evaluation results, the preliminary education business plan is optimized and adjusted to obtain the target education business plan.

9. The education business processing method based on big data as described in claim 8, characterized in that, The step of optimizing and adjusting the preliminary education business plan based on the evaluation results to obtain the target education business plan includes: Root cause analysis is performed on the deviations in the evaluation results to obtain a deviation tracing report for the educational operations; Based on the deviation tracing report, the parameter configuration of the preliminary education business plan is reconstructed and corrected to obtain the reconstructed education business plan; The constraint compliance of the reconstruction scheme is verified to obtain the verification conclusion of the education business. When the verification conclusions meet all constraints, the verified reconstruction scheme will be determined as the target education business scheme for the education business.

10. A big data-based education business processing system, characterized in that: The system includes: The data standardization and integration module is used to standardize and integrate the raw educational data from the education management platform and student learning terminals to obtain standardized educational data for educational business. The targeted feature extraction module is used to extract targeted features from the standardized education data based on the business objectives of the education business, so as to obtain a multi-dimensional business feature set of the education business. The correlation analysis module is used to perform correlation analysis on the multi-dimensional business feature set to obtain the business indicators of the education business. The business logic association module is used to associate historical education business processing data with the business indicators to obtain a preliminary education business plan. The decision framework construction module is used to generate a multi-objective optimization decision framework for the education business based on the key constraints of the education business. The solution iteration and optimization module is used to iteratively improve the preliminary education business solution based on the multi-objective optimization decision framework to obtain the target education business solution.

Images