Intelligent management and application method and system for digital teaching resource library

Abstract

The invention relates to the technical field of education informatization, and discloses a digital teaching resource library intelligent management and application method and system. The method comprises the steps of constructing a digital resource library, constructing a resource application big data analysis model to calculate a resource quality index, a popularity index and an update index, performing intelligent resource sorting and elimination based on the indexes, analyzing the correlation between a user retrieval request and resources, and performing intelligent recommendation. According to the method, efficient management and accurate application of teaching resources are realized, the resource utilization rate and the teaching effect are improved, the storage space and the management cost are saved, and data support is provided for education decision making.

Description

Technical Field

[0001] The present application relates to the field of educational information technology, and in particular to the field of digital teaching resource management and application technology. Background Art

[0002] In recent years, with the rapid development of information technology and the continuous advancement of educational informatization, digital teaching resources have played an increasingly important role in modern education. Educators and learners are increasingly relying on rich and diverse digital teaching resources to support the teaching and learning process. Therefore, building a high-quality massive teaching resource library that integrates resource collection, digitization, classification integration, management and evaluation to meet the needs of large-scale teaching resource applications has become an important topic in the field of education.

[0003] However, in practical applications, the construction and management of traditional digital teaching resource libraries face many challenges. First, with the sharp increase in the number of resources, there are a lot of problems with irregular content classification and identification in the resource library. This makes resource management, retrieval and use difficult, which not only reduces the user experience, but also causes a waste of storage space. Secondly, the quality of teaching resources varies. High-quality resources are difficult to be effectively identified and promoted, while low-quality resources take up a lot of space. In addition, the resource update speed is slow, and it is unable to reflect the latest teaching needs and subject development in a timely manner, resulting in reduced timeliness and applicability of resources.

[0004] Another significant problem is the imperfect evaluation and feedback mechanism of resource application effects. Traditional resource management systems often lack the ability to monitor and analyze resource usage in real time, and cannot accurately grasp the actual application effects of resources and user needs. This makes the optimization and update of resource libraries lack strong data support, making it difficult to achieve accurate resource recommendations and personalized services.

[0005] At the same time, the intelligent matching and recommendation of teaching resources also face challenges. The existing recommendation mechanism is often too simple and cannot fully consider the quality, popularity, update status of resources and the personalized needs of users, resulting in inaccurate recommendation results and failure to meet the diverse needs of educators and learners.

[0006] In addition, the lack of dynamic optimization and self-improvement capabilities of resource libraries is also a prominent problem. Many resource libraries lack an effective resource elimination mechanism, resulting in outdated and low-quality resources occupying storage space for a long time. At the same time, resource library updates often rely on manual operations, which are inefficient and prone to errors, making it difficult to respond to rapid changes in the education field in a timely manner.

[0007] These problems have seriously affected the effective use of digital teaching resources and restricted the in-depth development of educational informatization. Therefore, there is an urgent need for a new method that can achieve intelligent management, dynamic optimization, and accurate recommendation of teaching resources to improve resource utilization efficiency, enhance teaching effectiveness, and provide strong support for educational decision-making. Summary of the invention

[0008] The purpose of this application is to provide a method and system for intelligent management and application of a digital teaching resource library to solve the problems raised in the above background technology.

[0009] The present application discloses a method for intelligent management and application of a digital teaching resource library, comprising the following steps:

[0010] Step 100: constructing a digital resource library, including collecting teaching resources, and digitally processing and storing the teaching resources;

[0011] Step 200: construct a resource application big data analysis model, which is based on the resource application behavior data in the digital resource library and calculates the following indicators for subsequent resource management and application: a) resource quality index, including resource content applicability, resource content rationality and user feedback data; b) resource popularity index, including resource application trend degree, resource application frequency degree and user application feedback data; c) resource update index, including resource update time, update degree and update effect;

[0012] Step 300: Based on the resource quality index, resource heat index and resource update index obtained in step 200, intelligently sort the resources in the digital resource library to obtain a resource sorting result;

[0013] Step 400: According to the resource sorting result of step 300, in combination with the preset resource quality index threshold, resource heat index threshold and resource update index threshold, the resources at the end of the sorting and whose three indexes are all lower than the corresponding thresholds are intelligently eliminated;

[0014] Step 500: receiving a resource search request from a user, and analyzing the relevance between the user's search statement and each resource in the digital resource library;

[0015] Step 600: Based on the correlation analysis result of step 500 and the resource ranking result of step 300, intelligently recommend resources.

[0016] In a preferred example, the calculation formula of the resource quality index F(x) is:

[0017] F(x)=α1*(F(x) (1) )+α2*(F(x) (2) )+α3*(F(x) (3)) (1)

[0018] in:

[0019] F(x) (1) Indicates the applicability of resource content, reflecting the degree of match between resource content and teaching objectives, including the number of knowledge points covered by the resource, the matching between the requirements for mastering knowledge points and the level of detail of resource content;

[0020] F(x) (2) Indicates the rationality of resource content, reflecting the matching degree between resource content and the teaching stage and the cognitive level of the teaching group, including the matching between the mastery requirements of each knowledge point in the resource and the average cognitive level of the corresponding knowledge point of the teaching group at this teaching stage;

[0021] F(x) (3) It represents the user feedback results, reflecting the evaluation of teachers and students on the quality of resources, including the frequency of resource application and the proportion of resources that are fed back as high-quality resources;

[0022] α1, α2, and α3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource quality assessment.

[0023] In a preferred example, the calculation of the resource quality index F(x) also includes the following specific formula:

[0024]

[0025] F(x) (2) =∈1Match{(G1), (R1)}+∈2Match{(G2), (R′2)}+…+∈ m Match{(G m ′), (R m )} (3)

[0026]

[0027] in:

[0028] For formula (2):

[0029] Min(a, b) represents the total number of knowledge points covered by the resource

[0030] Max(a, b) represents the total number of knowledge points that should be mastered in this teaching stage under the teaching objectives

[0031] E c Indicates the requirements for mastering the cth subject knowledge point

[0032] F c Indicates the level of detail of the resource's explanation of the cth topic knowledge point

[0033] E d Indicates the mastery requirement of the dth general knowledge point

[0034] F d Indicates the level of detail of the resource's explanation of the dth general knowledge point

[0035] The Match{} function indicates the degree of match between two parameters.

[0036] ε1, ε2, ε3 are weight coefficients

[0037] For formula (3):

[0038] G k (k=1, 2, ..., m) represents the mastery requirement of the kth knowledge point contained in the resource at this teaching stage

[0039] R k '(k=1,2,...,m) represents the average cognitive level of the teaching group on the kth knowledge point in this teaching stage

[0040] ∈ k (k=1, 2, ..., m) is the weight coefficient

[0041] For formula (4):

[0042] K (apply this resource) indicates the total number of times this resource is used by teachers in teaching.

[0043] K (Feedback on this resource) indicates the total number of times teachers give feedback that this resource is a high-quality resource after using it for teaching.

[0044] L (应用该资源) Indicates the total number of times the resource has been used by students for learning

[0045] L (反馈该资源) Indicates the total number of times students give feedback that the resource is a high-quality resource after applying it.

[0046] K (n) and L n Represents the total number of times the nth resource is used by teachers and students respectively

[0047] N represents the total number of resources contained in the sub-library to which the resource belongs

[0048] μ1, μ2, μ3, μ4 are weight coefficients

[0049] These formulas together constitute the detailed calculation method of the resource quality index F(x), which comprehensively evaluates the content applicability, content rationality and user feedback of the resource.

[0050] In a preferred example, the calculation formula of the resource heat index F(y) is:

[0051] F(y)=β1*(F(y) (1) )+β2*(F(y) (2) )+β3*(F(y) (3) ) (5)

[0052] in:

[0053] F(y) (1) It indicates the degree of resource application trend, reflecting the application trend of resources in a certain period of time, including the number, frequency and time distribution of teachers' application teaching and students' application learning;

[0054] F(y) (2) Indicates the frequency of resource application, reflecting the overall situation of resource use by teachers and students, including the total number of times the resource is used, the total duration, and the proportion of its application in all resources;

[0055] F(y) (3) It indicates the user application feedback results, reflecting the direct feedback of teachers and students after applying the resources, including the proportion of effective feedback behaviors such as likes, collections, and comments;

[0056] β1, β2, and β3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource heat assessment.

[0057] In a preferred example, the calculation of the resource heat index F(y) also includes the following specific formula:

[0058]

[0059]

[0060]

[0061] in:

[0062] For formula (6):

[0063] T 分析 Indicates the interval between two resource heat analyses

[0064] d ku Indicates the teacher's application behavior of the resource on the kth day uth time

[0065] e kv Indicates the student's application behavior for the resource on the kth day and the vth time

[0066] s 1u Indicates the weight value of the teacher's application behavior for the uth time of this resource

[0067] s 2vIndicates the weight value of the v-th student application behavior of this resource

[0068] Indicates the date weight value of the kth day

[0069] u′ represents the total number of times the resource is used by teachers for teaching on the kth day

[0070] v′ represents the total number of times the resource is used by students for learning on the kth day

[0071] π1 is the weight coefficient

[0072] For formula (7):

[0073] b ih Indicates that the resource is ih Teachers h times applied

[0074] t ih Indicates that the resource is ih The application time of the hth application by each teacher

[0075] c jg Indicates that the resource is used by the jth student for the gth time

[0076] t jg Indicates the time when the resource was used by the jth student for the gth time

[0077] i′ represents the total number of teachers who apply all resources

[0078] h′ represents the total number of times the resource is used by teachers in teaching

[0079] j′ represents the total number of students who applied all resources

[0080] g′ represents the total number of times the resource is used by students for learning

[0081] ρ1, ρ2 are weight coefficients

[0082] For formula (8):

[0083] P s Indicates the total number of times the resource has been used by teachers for teaching

[0084] P′ s Indicates the number of times the teacher provides effective feedback after using the resource in teaching

[0085] Q r Indicates the total number of times this resource has been used by students for learning

[0086] Q′ r Indicates the number of times students provide effective feedback after applying the resource.

[0087] σ1, σ1 are weight coefficients

[0088] These formulas together constitute the detailed calculation method of the resource heat index F(y), which comprehensively evaluates the application trend, application frequency and user feedback of the resource.

[0089] In a preferred example, the calculation formula of the resource update index F(z) is:

[0090] F(z)=γ1*(F(z) (1) )+γ2*(F(z) (2) )+γ3*(F(z) (3) ) (9)

[0091] in:

[0092] F(z) (1) Indicates resource update time, reflecting the timeliness of resources, including the relationship between resource storage time, latest update time and current time;

[0093] F(z) (2) Indicates the resource update degree, reflecting the amount of change in resource content, including the number of resource updates within a certain time interval, and the ratio of the updated data volume to the original data volume;

[0094] F(z) (3) Indicates the effect of user feedback on the update, reflecting the evaluation of teachers and students on the effect of the resource update, including the proportion of updated resources that were fed back as having significant effects;

[0095] γ1, γ2, and γ3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource renewal assessment;

[0096] Among them, ε1, ε2, ε3, β1, β2, β3, γ1, γ2, and γ3 are all weight coefficients, which are used to adjust the importance of each factor in the calculation of resource quality index, resource heat index, and resource update index. These weight coefficients can be adjusted according to specific application scenarios and needs.

[0097] In a preferred example, the calculation of the resource update index F(z) also includes the following specific formula:

[0098]

[0099]

[0100]

[0101] in:

[0102] For formula (10):

[0103] T统计2 Indicates the time when the system is currently performing resource analysis

[0104] T 统计1 Indicates the time when the system last performed resource analysis

[0105] T 入库 Indicates the time when the resource was put into storage

[0106] T 更新 Indicates the latest update time of the resource

[0107] ω1, ω2 are weight coefficients

[0108] For formula (11):

[0109] S 更新 Indicates the number of times the resource is updated between two resource analysis intervals

[0110] S′ 更新 Indicates the sum of data updates between two resource analysis intervals

[0111] S′ 原始 Indicates the total amount of raw data

[0112] w represents the adjustment coefficient

[0113] is the weight coefficient

[0114] For formula (12):

[0115] R′ p Indicates the number of times the resource has been updated with significant feedback from teachers after being used in teaching

[0116] R p Indicates the teacher's feedback result on the resource data update for the pth time

[0117] p′ represents the total number of times teachers provide feedback on the effectiveness of resource updates

[0118] W′ q Indicates the number of times the resource has been updated with significant feedback from students after application and learning.

[0119] W q Indicates the feedback result of the student's qth update of the resource data

[0120] q′ represents the total number of times students provide feedback on the effect of resource updating

[0121] τ1, τ2 are weight coefficients

[0122] These formulas together constitute the detailed calculation method of the resource update index F(z), which comprehensively evaluates the update time, update degree and update effect of the resource. Formula (10) reflects the update time of the resource, formula (11) measures the update degree of the resource, and formula (12) evaluates the user's feedback on the resource update effect.

[0123] In a preferred example, the calculation formula of the resource ranking result F1 in step 300 is:

[0124] F1=δ1*F(x)+δ2*F(y)+δ3*F(z)(13)

[0125] in:

[0126] F(x) is the resource quality index calculated in step 200, reflecting the content quality of the resource;

[0127] F(y) is the resource heat index calculated in step 200, reflecting the popularity of the resource;

[0128] F(z) is the resource update index calculated in step 200, reflecting the timeliness of the resource;

[0129] δ1, δ2, and δ3 are weight coefficients used to adjust the importance of resource quality, popularity, and update in resource sorting. These weight coefficients can be adjusted according to specific application scenarios and requirements;

[0130] The resource ranking result F1 comprehensively considers the quality, popularity and timeliness of the resources, and is used to comprehensively evaluate and sort the resources in the digital resource library. The higher the F1 value, the higher the comprehensive value of the resource, and the higher the ranking in the resource library.

[0131] In a preferred example, the step 500 specifically includes: analyzing the matching degree between the user search statement and each resource in the digital resource library, wherein the matching degree is selected from the following group: keyword matching degree, synonym matching degree, vocabulary substring or parent string matching degree, core semantic matching degree and resource encoding matching degree; and

[0132] The step 600 specifically includes: calculating the resource recommendation degree based on the matching degree analysis result of step 500 and the resource ranking result of step 300, and intelligently recommending resources according to the resource recommendation degree; and,

[0133] The calculation formula of the resource recommendation degree F2 in step 600 is:

[0134]

[0135] in:

[0136] F′ represents the keyword matching degree between the user's search statement and the resource;

[0137] F″ represents the synonym matching degree between the user’s search statement and the resource;

[0138] F″′ represents the degree of matching between the user’s search statement and the resource’s vocabulary substring or parent string;

[0139] F″″ represents the core semantic matching degree between the user’s search statement and the resource;

[0140] F° represents the matching degree between the user’s search code and the resource code;

[0141] F1 is the resource sorting result obtained in step 300;

[0142] k1 and k2 are weight coefficients, which are used to adjust the importance of matching analysis results and resource ranking results in the recommendation calculation;

[0143] is the weight coefficient, which is used to adjust the importance of semantic matching and encoding matching in the recommendation calculation;

[0144] θ1, θ2, θ3, and θ4 are weight coefficients used to adjust the importance of different types of matching in semantic matching;

[0145] The resource recommendation degree F2 comprehensively considers the matching degree between the user's search requirements and the resources and the comprehensive ranking results of the resources, and is used to determine the priority of recommending resources to the user. The higher the F2 value, the more the resource meets the user's needs and the higher the comprehensive value, and the more priority it is recommended to the user.

[0146] The present application also discloses a digital teaching resource library intelligent management and application system, including:

[0147] A digital resource library construction module is used to collect teaching resources, digitally process and store the teaching resources, and construct a digital resource library;

[0148] The resource application big data analysis module is used to construct a resource application big data analysis model based on the resource application behavior data in the digital resource library, and calculate the following indicators for subsequent resource management and application: a) resource quality index, including resource content applicability, resource content rationality and user feedback data; b) resource popularity index, including resource application trend degree, resource application frequency degree and user application feedback data; c) resource update index, including resource update time, update degree and update effect;

[0149] A resource intelligent sorting module, used to intelligently sort the resources in the digital resource library based on the resource quality index, resource heat index and resource update index obtained by the resource application big data analysis module to obtain a resource sorting result;

[0150] A resource intelligent elimination module is used to intelligently eliminate resources at the end of the sorting and whose three indexes are all lower than the corresponding thresholds according to the resource sorting results of the resource intelligent sorting module and in combination with a preset resource quality index threshold, a resource heat index threshold and a resource update index threshold;

[0151] A resource search analysis module, used to receive a user's resource search request and analyze the relevance between the user's search statement and each resource in the digital resource library;

[0152] The resource intelligent recommendation module is used to make intelligent recommendations on resources based on the correlation analysis results of the resource retrieval analysis module and the resource sorting results of the resource intelligent sorting module.

[0153] The implementation methods of this application have the following technical effects:

[0154] The digital teaching resource library intelligent management and application method and system of this application has brought significant technical effects in the field of educational resource management and application. First of all, it realizes the efficient management and precise application of teaching resources. Through the comprehensive evaluation of resource quality index, heat index and update index, the method conducts a comprehensive quantitative evaluation of resources. Based on these evaluation results, the system can perform intelligent sorting and elimination, effectively ensuring the high quality of resources in the resource library. At the same time, the intelligent matching and recommendation functions greatly improve the accuracy and efficiency of resource retrieval, allowing educators and learners to find the required high-quality resources more quickly.

[0155] Secondly, this application significantly improves the dynamic optimization capabilities of the resource library. The resource application big data analysis model realizes real-time monitoring of resource usage, and the resource library can be continuously updated and optimized through the feedback mechanism of resource elimination and matching information. This mechanism not only ensures the timeliness of the resource library content, but also maintains the diversity of resource themes, so that the resource library can always meet the latest needs of teaching and learning.

[0156] In terms of teaching effectiveness, the application of this application improves the utilization rate of teaching resources and the quality of teaching. The intelligent recommendation of high-quality resources helps teachers choose the most suitable teaching materials, thereby improving the quality of teaching. The precise matching of resources meets the personalized needs of different users. Whether it is teachers preparing lessons or students learning independently, they can find the most suitable resources. The continuous optimization of the resource library ensures that the teaching content keeps pace with the times, enabling education to keep up with the pace of social development and scientific and technological progress.

[0157] From a management perspective, this application saves a lot of storage space and management costs. The intelligent elimination mechanism reduces the occupation of redundant and obsolete resources, and the automated management process greatly reduces the cost of manual maintenance. The efficient use of resources also improves the performance of the overall system, allowing limited hardware resources to support more users' simultaneous access and use.

[0158] Finally, this application provides important data support for educational decision-making. By analyzing resource application data, educational administrators and teachers can obtain valuable evidence for improving teaching content and methods. User feedback data helps to understand learners' needs and learning effects, so as to develop more targeted teaching strategies and learning plans.

[0159] In summary, the intelligent management and application method and system of the digital teaching resource library of this application realizes efficient management and precise application of teaching resources through intelligent means, providing strong support for improving teaching quality and learning effects.

[0160] A large number of technical features are recorded in the specification of this application, which are distributed in various technical solutions. If all possible combinations of technical features of this application (i.e., technical solutions) are to be listed, the specification will be too long. In order to avoid this problem, the various technical features disclosed in the above-mentioned invention content of this application, the various technical features disclosed in the various embodiments and examples below, and the various technical features disclosed in the accompanying drawings can be freely combined with each other to form various new technical solutions (these technical solutions are all deemed to have been recorded in this specification), unless the combination of such technical features is technically infeasible. For example, in one example, feature A+B+C is disclosed, and in another example, feature A+B+D+E is disclosed, and features C and D are equivalent technical means that play the same role. Technically, only one can be used, and it is impossible to use them at the same time. Feature E can be combined with feature C technically. Then, the solution of A+B+C+D should not be deemed to have been recorded because it is technically infeasible, and the solution of A+B+C+E should be deemed to have been recorded. BRIEF DESCRIPTION OF THE DRAWINGS

[0161] Figure 1 It is a system design block diagram of the intelligent management and application method of the digital teaching resource library according to this application.

[0162] Figure 2 It is a flow chart of the intelligent management and application method of the digital teaching resource library according to the first implementation mode of the present application.

[0163] Figure 3 It is a structural diagram of a digital teaching resource library intelligent management and application system according to the second implementation mode of the present application. DETAILED DESCRIPTION

[0164] In the following description, many technical details are provided to help readers better understand the present application. However, those skilled in the art can understand that the technical solution claimed in the present application can be implemented even without these technical details and various changes and modifications based on the following embodiments.

[0165] Description of some concepts:

[0166] Digital teaching resources: refers to various materials in digital form used to support teaching and learning activities, including but not limited to electronic textbooks, courseware, videos, audios, pictures, test questions, etc.

[0167] Resource quality index: In this application, it refers to a quantitative indicator used to measure the quality of teaching resources, which comprehensively considers the applicability of resource content, rationality of resource content and user feedback data.

[0168] Resource heat index: In this application, it refers to a quantitative indicator used to measure the popularity or attention of a resource, which comprehensively considers the resource application trend, resource application frequency and user application feedback data.

[0169] Resource Update Index: In this application, it refers to a quantitative indicator used to measure the degree of resource update, which comprehensively considers the resource update time, update degree and update effect.

[0170] Resource application big data analysis model: In this application, it refers to a mathematical model that calculates resource quality index, popularity index and update index based on resource application behavior data in the resource library.

[0171] Intelligent resource sorting: In this application, it refers to the process of comprehensively evaluating and sorting resources in the resource library based on resource quality index, popularity index and update index.

[0172] Intelligent resource elimination: In this application, it refers to the process of automatically identifying and removing low-quality, low-popularity, and long-unupdated resources based on resource sorting results and preset thresholds.

[0173] Intelligent resource matching: In this application, it refers to the process of analyzing the relevance of user search statements and various resources in the resource library to find the resources that best meet user needs.

[0174] Intelligent resource recommendation: In this application, it refers to the process of recommending the most suitable teaching resources to users based on resource matching and resource ranking results.

[0175] Resource code: In this application, it refers to a digital unique identifier based on resource attribute information and content feature information, which is used for rapid classification, storage and retrieval of resources.

[0176] Resource management information: In this application, it refers to the summary information of resource application in each application system, which is used for resource application behavior data analysis.

[0177] Resource content feature information: In this application, it refers to the important content feature description information related to resources and teaching applications, such as keywords, subject knowledge points, etc., which are used for resource retrieval and recommendation.

[0178] Resource content applicability: In this application, it refers to an indicator that reflects the degree of match between resource content and teaching objectives, including the number of knowledge points covered by the resource, the matching of knowledge point mastery requirements and the level of detail of the resource content.

[0179] Resource content rationality: In this application, it refers to an indicator that reflects the degree of match between resource content and the teaching stage and cognitive level of the teaching group.

[0180] Resource application trend degree: In this application, it refers to an indicator that reflects the application trend of resources within a certain period of time.

[0181] Frequency of resource application: In this application, it refers to an indicator that reflects the overall situation of resource use by teachers and students.

[0182] Effective feedback behavior: In this application, it refers to the user's active interaction with resources, such as likes, favorites, comments, etc.

[0183] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below in conjunction with the accompanying drawings.

[0184] The first embodiment of the present application relates to a method for intelligent management and application of a digital teaching resource library, and its system design diagram is as follows: Figure 1 As shown, the process is as follows Figure 2 As shown, the method comprises the following steps:

[0185] Step 100: constructing a digital resource library, including collecting teaching resources, and digitally processing and storing the teaching resources.

[0186] Optionally, the step 100 specifically includes: digitally encoding, deduplicating and classifying and storing the teaching resources, and constructing a digital resource library including resource codes, resource management information, resource content feature information and resource content.

[0187] Specifically, building a digital resource library is the basic work of the entire system, involving multiple key technical links. First, the system needs to collect teaching resources through various channels, which may include web crawler technology, API interface docking, or user uploads. The collected original resources may exist in multiple formats, such as text, pictures, audio, video, etc., so unified digital processing is required.

[0188] Digital processing is a complex technical process, including format conversion, content extraction, metadata generation and other steps. For example, for text resources, OCR (optical character recognition) processing may be required; for audio and video resources, codec conversion and key frame extraction may be required. The goal of this process is to convert all resources into a digital format that can be uniformly managed and processed by the system.

[0189] The next key step is resource coding. The system generates a unique digital code for each resource. This code is not only a unique identifier for the resource, but may also contain key information such as the resource type, source, and creation time. The design of the code needs to consider scalability and retrieval efficiency, and usually adopts a multi-level coding structure.

[0190] Deduplication is an important step in ensuring the quality of the resource library. The system uses hash algorithms, content feature comparison and other technical means to identify and merge duplicate resources. This process not only saves storage space, but also improves the accuracy of resource retrieval.

[0191] Classification storage is the key to optimizing resource management and retrieval efficiency. The system will automatically classify resources based on their content features, metadata and other information using machine learning algorithms (such as support vector machines, neural networks, etc.). The design of the classification system needs to take into account the particularities of the education field and may adopt a multi-dimensional, hierarchical classification structure.

[0192] Finally, the system will store the processed resources together with their coding, management information (such as access rights, usage statistics, etc.), and content feature information (such as keywords, themes, difficulty levels, etc.) in the database. Considering the diversity and complexity of educational resources, a hybrid storage strategy may be adopted here, such as using a relational database to store structured metadata information, and using object storage or distributed file systems to store large-capacity resource content files.

[0193] This carefully designed digital resource library construction process lays a solid technical foundation for subsequent intelligent management and application, ensuring efficient storage, rapid retrieval and accurate recommendation of resources.

[0194] Step 200: Construct a resource application big data analysis model, which is based on the resource application behavior data in the digital resource library and calculates the following indicators for subsequent resource management and application: a) resource quality index, including resource content applicability, resource content rationality and user feedback data; b) resource popularity index, including resource application trend degree, resource application frequency degree and user application feedback data; c) resource update index, including resource update time, update degree and update effect.

[0195] Specifically, the resource application big data analysis model constructed in step 200 is the core intelligent engine of the entire system. It uses advanced data analysis technology to extract valuable information from massive resource application behavior data and convert it into quantifiable indicators. This process involves complex data processing, statistical analysis and machine learning algorithms.

[0196] First, the system needs to collect and preprocess a large amount of resource application behavior data. This data may come from multiple dimensions such as user browsing history, download history, usage time, ratings and comments. The data preprocessing stage includes steps such as data cleaning (removing outliers and noise data), data standardization, and feature extraction. This stage may use big data processing frameworks such as ApacheSpark or Hadoop to process massive data.

[0197] Next, the system constructs three key indicators based on the processed data: resource quality index, resource heat index and resource update index. The calculation of each indicator involves the comprehensive evaluation of multiple sub-indicators.

[0198] For the resource quality index, the system analyzes resource content applicability, content rationality, and user feedback data. The calculation of content applicability may involve natural language processing (NLP) techniques, such as semantic analysis and topic modeling, to evaluate the degree of match between resource content and teaching objectives. The evaluation of content rationality may use machine learning algorithms to analyze the match between resource difficulty and the cognitive level of the target user group. User feedback data may be processed through sentiment analysis algorithms to process user comments and combined with rating data for comprehensive evaluation.

[0199] The calculation of resource heat index involves time series analysis and trend prediction techniques. The system may use moving average, exponential smoothing, or more complex ARIMA models to analyze resource application trends. The calculation of application frequency may consider the time decay factor, that is, the most recent application behavior has a higher weight. The processing of user application feedback data may use collaborative filtering or matrix decomposition and other commonly used techniques in recommendation systems.

[0200] The calculation of resource update index requires tracking the version history and update records of resources. The evaluation of update time may use a time decay function to give more weight to the most recent updates. The calculation of update degree may involve text similarity analysis to compare the differences between different versions. The evaluation of update effect may be carried out in combination with user feedback and changes in resource usage.

[0201] In the calculation process of these three indicators, the system may use a variety of machine learning models, such as random forests, gradient boosting trees, or deep learning models, to integrate the various sub-indicators and derive the final comprehensive score. At the same time, the system may use technologies such as A / B testing or multi-armed bandit algorithms to continuously optimize the calculation weights of each indicator.

[0202] Finally, this big data analysis model is not static, but will continue to learn and adjust. The system may use online learning algorithms to continuously update model parameters based on new application data and feedback to adapt to the dynamic changes in educational resource usage patterns.

[0203] Through this complex big data analysis model, the system can comprehensively and dynamically evaluate the quality, popularity and timeliness of each educational resource, providing strong data support for subsequent resource management and personalized recommendations.

[0204] Optionally, the calculation formula of the resource quality index F(x) is:

[0205] F(x)=α1*(F(x) (1) )+α2*(F(x) (2) )+α3*(F(x) (3) ) (1)

[0206] in:

[0207] F(x) (1) Indicates the applicability of resource content, reflecting the degree of match between resource content and teaching objectives, including the number of knowledge points covered by the resource, the matching between the requirements for mastering knowledge points and the level of detail of resource content;

[0208] F(x) (2) Indicates the rationality of resource content, reflecting the matching degree between resource content and the teaching stage and the cognitive level of the teaching group, including the matching between the mastery requirements of each knowledge point in the resource and the average cognitive level of the corresponding knowledge point of the teaching group at this teaching stage;

[0209] F(x) (3) It represents the user feedback results, reflecting the evaluation of teachers and students on the quality of resources, including the frequency of resource application and the proportion of resources that are fed back as high-quality resources;

[0210] α1, α2, and α3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource quality assessment.

[0211] Optionally, the calculation of the resource quality index F(x) also includes the following specific formula:

[0212]

[0213] F(x) (2) =∈1Match{(G1), (R1)}+∈2Match{(G2), (R′2)}+…+∈ m Match{(G m ′), (R m )} (3)

[0214]

[0215] in:

[0216] For formula (2):

[0217] Min(a, b) represents the total number of knowledge points covered by the resource

[0218] Max(a, b) represents the total number of knowledge points that should be mastered in this teaching stage under the teaching objectives

[0219] E c Indicates the requirements for mastering the cth subject knowledge point

[0220] F c Indicates the level of detail of the resource's explanation of the cth topic knowledge point

[0221] E d Indicates the mastery requirement of the dth general knowledge point

[0222] F d Indicates the level of detail of the resource's explanation of the dth general knowledge point

[0223] The Match{} function indicates the degree of match between two parameters.

[0224] ε1, ε2, ε3 are weight coefficients

[0225] For formula (3):

[0226] G k (k=1, 2, ..., m) represents the mastery requirement of the kth knowledge point contained in the resource at this teaching stage

[0227] R k '(k=1,2,...,m) represents the average cognitive level of the teaching group on the kth knowledge point in this teaching stage

[0228] ∈ k (k=1, 2, ..., m) is the weight coefficient

[0229] For formula (4):

[0230] K (apply this resource) indicates the total number of times this resource is used by teachers in teaching.

[0231] K (Feedback on this resource) indicates the total number of times teachers give feedback that this resource is a high-quality resource after using it for teaching.

[0232] L (应用该资源) Indicates the total number of times the resource has been used by students for learning

[0233] L (反馈该资源) Indicates the total number of times students give feedback that the resource is a high-quality resource after applying it.

[0234] K (n) and L n Represents the total number of times the nth resource is used by teachers and students respectively

[0235] N represents the total number of resources contained in the sub-library to which the resource belongs

[0236] μ1, μ2, μ3, μ4 are weight coefficients

[0237] These formulas together constitute the detailed calculation method of the resource quality index F(x), which comprehensively evaluates the content applicability, content rationality and user feedback of the resource.

[0238] Optionally, the calculation formula of the resource heat index F(y) is:

[0239] F(y)=β1*(F(y) (1) )+β2*(F(y) (2) )+β3*(F(y) (3) ) (5)

[0240] in:

[0241] F(y) (1) It indicates the degree of resource application trend, reflecting the application trend of resources in a certain period of time, including the number, frequency and time distribution of teachers' application teaching and students' application learning;

[0242] F(y) (2) Indicates the frequency of resource application, reflecting the overall situation of resource use by teachers and students, including the total number of times the resource is used, the total duration, and the proportion of its application in all resources;

[0243] F(y) (3)It indicates the user application feedback results, reflecting the direct feedback of teachers and students after applying the resources, including the proportion of effective feedback behaviors such as likes, collections, and comments;

[0244] β1, β2, and β3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource heat assessment.

[0245] Optionally, the calculation of the resource heat index F(y) also includes the following specific formula:

[0246]

[0247]

[0248]

[0249] in:

[0250] For formula (6):

[0251] T 分析 Indicates the interval between two resource heat analyses

[0252] d ku Indicates the teacher's application behavior of the resource on the kth day uth time

[0253] e kv Indicates the student's application behavior for the resource on the kth day and the vth time

[0254] s 1u Indicates the weight value of the teacher's application behavior for the uth time of this resource

[0255] s 2v Indicates the weight value of the v-th student application behavior of this resource

[0256] Indicates the date weight value of the kth day

[0257] u′ represents the total number of times the resource is used by teachers for teaching on the kth day

[0258] v′ represents the total number of times the resource is used by students for learning on the kth day

[0259] π1 is the weight coefficient

[0260] For formula (7):

[0261] b ih Indicates that the resource is ih Teachers h times applied

[0262] t ih Indicates that the resource is ih The application time of the hth application by each teacher

[0263] c jg Indicates that the resource is used by the jth student for the gth time

[0264] t jg Indicates the time when the resource was used by the jth student for the gth time

[0265] i′ represents the total number of teachers who apply all resources

[0266] h′ represents the total number of times the resource is used by teachers in teaching

[0267] j′ represents the total number of students who applied all resources

[0268] g′ represents the total number of times the resource is used by students for learning

[0269] ρ1, ρ2 are weight coefficients

[0270] For formula (8):

[0271] P s Indicates the total number of times the resource has been used by teachers for teaching

[0272] P′ s Indicates the number of times the teacher provides effective feedback after using the resource in teaching

[0273] Q r Indicates the total number of times this resource has been used by students for learning

[0274] Q′ r Indicates the number of times students provide effective feedback after applying the resource.

[0275] σ1, σ1 are weight coefficients

[0276] These formulas together constitute the detailed calculation method of the resource heat index F(y), which comprehensively evaluates the application trend, application frequency and user feedback of the resource.

[0277] Optionally, the calculation formula of the resource update index F(z) is:

[0278] F(z)=γ1*(F(z) (1) )+γ2*(F(z) (2) )+γ3*(F(z) (3) ) (9)

[0279] in:

[0280] F(z) (1) Indicates resource update time, reflecting the timeliness of resources, including the relationship between resource storage time, latest update time and current time;

[0281] F(z) (z) Indicates the resource update degree, reflecting the amount of change in resource content, including the number of resource updates within a certain time interval, and the ratio of the updated data volume to the original data volume;

[0282] F(z) (3) Indicates the effect of user feedback on the update, reflecting the evaluation of teachers and students on the effect of the resource update, including the proportion of updated resources that were fed back as having significant effects;

[0283] γ1, γ2, and γ3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource renewal assessment;

[0284] Among them, ε1, ε2, ε3, β1, β2, β3, γ1, γ2, and γ3 are all weight coefficients, which are used to adjust the importance of each factor in the calculation of resource quality index, resource heat index, and resource update index. These weight coefficients can be adjusted according to specific application scenarios and needs.

[0285] Optionally, the calculation of the resource update index F(z) also includes the following specific formula:

[0286]

[0287]

[0288]

[0289] in:

[0290] For formula (10):

[0291] T 统计2 Indicates the time when the system is currently performing resource analysis

[0292] T 统计1 Indicates the time when the system last performed resource analysis

[0293] T 入库 Indicates the time when the resource was put into storage

[0294] T 更新 Indicates the latest update time of the resource

[0295] ω1, ω2 are weight coefficients

[0296] For formula (11):

[0297] S 更新 Indicates the number of times the resource is updated between two resource analysis intervals

[0298] S′ 更新Indicates the sum of data updates between two resource analysis intervals

[0299] S′ 原始 Indicates the total amount of raw data

[0300] w represents the adjustment coefficient

[0301] is the weight coefficient

[0302] For formula (12):

[0303] R′ p Indicates the number of times the resource has been updated with significant feedback from teachers after being used in teaching

[0304] R p Indicates the teacher's feedback result on the resource data update for the pth time

[0305] p′ represents the total number of times teachers provide feedback on the effectiveness of resource updates

[0306] W′ q Indicates the number of times the resource has been updated with significant feedback from students after application and learning.

[0307] W q Indicates the feedback result of the student's qth update of the resource data

[0308] q′ represents the total number of times students provide feedback on the effect of resource updating

[0309] τ1, τ2 are weight coefficients

[0310] These formulas together constitute the detailed calculation method of the resource update index F(z), which comprehensively evaluates the update time, update degree and update effect of the resource. Formula (10) reflects the update time of the resource, formula (11) measures the update degree of the resource, and formula (12) evaluates the user's feedback on the resource update effect.

[0311] Step 300: Based on the resource quality index, resource heat index and resource update index obtained in step 200, the resources in the digital resource library are intelligently sorted to obtain a resource sorting result.

[0312] Specifically, step 300 is the key link of the intelligent management system of the digital teaching resource library, which converts the multi-dimensional indicators calculated in step 200 into actual resource sorting results. This process not only needs to comprehensively consider the resource quality index, popularity index and update index, but also needs to achieve dynamic and personalized sorting through complex algorithms.

[0313] First, the system needs to design a comprehensive scoring function. This function usually takes the form of weighted summation, that is, F = w1*quality index + w2*heat index + w3*update index, where w1, w2, w3 are the corresponding weight coefficients. Determining these weights is a key technical challenge, and multiple approaches may be used:

[0314] Expert system: Initial weights are set based on the advice of education experts.

[0315] Machine learning methods: Use historical data to train models, such as linear regression or support vector machines, to learn the optimal weight combination.

[0316] Reinforcement learning: Treat the ranking problem as a decision-making process and use algorithms such as Q-learning or policy gradient to dynamically adjust weights to maximize user satisfaction or other key performance indicators.

[0317] Genetic algorithm: It iteratively optimizes weight combinations by simulating the evolutionary process.

[0318] Once the scoring function is determined, the next step is to implement an efficient sorting algorithm. Considering that the repository may contain massive amounts of data, traditional sorting algorithms (such as quick sort) may not meet real-time requirements. Therefore, the system can adopt the following strategies:

[0319] Distributed sorting: Use distributed computing frameworks such as Apache Spark to distribute sorting tasks to multiple nodes for parallel processing.

[0320] Approximate sorting: For very large-scale data, approximate sorting algorithms such as locality sensitive hashing (LSH) or skip tables may be used to sacrifice a certain degree of accuracy in exchange for faster sorting speed.

[0321] Incremental sorting: Instead of re-sorting all resources every time, only the sorting position of the resources that have changed is updated. This can be achieved by maintaining an ordered index structure (such as a B+ tree).

[0322] Pre-calculation and caching: For common query scenarios, the system may pre-calculate and cache sorting results to improve response speed.

[0323] In addition, the system also needs to consider the personalization and contextual relevance of the sorting. This involves:

[0324] Collaborative filtering: Adjust the ranking results based on the user's historical behavior and the preferences of similar users.

[0325] Context-aware: Dynamically adjust the sorting weight based on the user’s current scenario (such as device type, time, location, etc.).

[0326] Diversity guarantee: Use algorithms such as Maximum Margin Relevance (MMR) to increase the diversity of results while ensuring relevance.

[0327] Finally, in order to continuously optimize the ranking effect, the system may implement an A / B testing framework to continuously try different ranking strategies and evaluate their effects. At the same time, an online learning mechanism may be introduced to dynamically adjust the ranking parameters based on real-time user feedback (such as click-through rate, dwell time, etc.).

[0328] Through this complex and sophisticated intelligent sorting process, the system can provide users with the most relevant and valuable educational resources, greatly improving resource utilization efficiency and user satisfaction.

[0329] Optionally, the calculation formula of the resource ranking result F1 in step 300 is:

[0330] F1=δ1*F(x)+δ2*F(y)+δ3*F(z)(13)

[0331] in:

[0332] F(x) is the resource quality index calculated in step 200, reflecting the content quality of the resource;

[0333] F(y) is the resource heat index calculated in step 200, reflecting the popularity of the resource;

[0334] F(z) is the resource update index calculated in step 200, reflecting the timeliness of the resource;

[0335] δ1, δ2, and δ3 are weight coefficients used to adjust the importance of resource quality, popularity, and update in resource sorting. These weight coefficients can be adjusted according to specific application scenarios and needs;

[0336] The resource ranking result F1 comprehensively considers the quality, popularity and timeliness of the resources, and is used to comprehensively evaluate and sort the resources in the digital resource library. The higher the F1 value, the higher the comprehensive value of the resource, and the higher the ranking in the resource library.

[0337] Step 400: Based on the resource sorting result of step 300, combined with the preset resource quality index threshold, resource heat index threshold and resource update index threshold, the resources at the end of the sorting and whose three indexes are all lower than the corresponding thresholds are intelligently eliminated.

[0338] Specifically, step 400 is the key optimization link of the intelligent management system of the digital teaching resource library, which maintains the quality and efficiency of the resource library through an intelligent elimination mechanism. This process not only relies on the sorting results in the previous steps, but also introduces multiple threshold judgments to ensure the accuracy and rationality of the elimination decision.

[0339] First, the system needs to determine the appropriate elimination thresholds. These thresholds include resource quality index thresholds, heat index thresholds, and update index thresholds, and their setting is a complex process:

[0340] Data analysis: Use statistical methods, such as quartile analysis or standard deviation, to determine initial thresholds based on historical data. For example, resources below the first quartile might be considered potential candidates for elimination.

[0341] Machine learning models: Classification models (such as decision trees or random forests) are trained to learn which feature combinations typically lead to resource elimination, thereby dynamically adjusting thresholds.

[0342] Time series analysis: Consider the historical trends of resource indicators, use models such as ARIMA or Prophet to predict future trends, and adjust thresholds accordingly.

[0343] Integration of expert knowledge: Incorporate the opinions of educational experts into the threshold setting process through methods such as the Delphi method.

[0344] After the threshold is determined, the system will make multiple judgments on the resources at the end of the sorting results:

[0345] Sorting position check: Identify a "danger zone", such as the last 20% of resources in the sorting process that may be considered potential candidates for elimination.

[0346] Multi-dimensional threshold comparison: For each resource in the "danger zone", compare its quality index, heat index, and update index to see if they are all lower than the corresponding threshold.

[0347] Comprehensive scoring: A comprehensive scoring function may be introduced, such as Score = w1*(Q-Qthreshold)+w2*(H-Hthreshold)+w3*(U-Uthreshold), where Q, H, and U represent quality, popularity, and update index, respectively. Elimination is considered only when the Score is less than a negative value.

[0348] Time window analysis: Consider the performance trend of resources in the recent period (such as the past 3 months) to avoid mistakenly eliminating potential resources due to short-term fluctuations.

[0349] To increase the robustness of the decision, the system may adopt the following strategies:

[0350] Soft elimination mechanism: Instead of directly deleting resources, they are marked as "pending elimination" and the final decision is made after a period of observation.

[0351] Gradual elimination: Set multiple elimination levels, such as "to be observed", "temporarily removed from shelves", "permanently deleted", etc., and choose different processing methods according to the specific circumstances of the resources.

[0352] Anomaly Detection: Use algorithms such as Isolation Forest or Autoencoders to identify and protect "anomalous" resources that may have unique value despite low metrics.

[0353] User feedback integration: Consider direct feedback from users in elimination decisions, such as retaining ratings or reviews even if other metrics are low.

[0354] Resource association analysis: Consider the association between resources to avoid affecting the integrity of other related resources due to the elimination of a certain resource.

[0355] In addition, the system also needs to consider the technical implementation of the elimination operation:

[0356] Distributed consistency: In a distributed system, ensuring the atomicity and consistency of elimination operations may use distributed transactions or consensus algorithms.

[0357] Backup and recovery: Establish a backup mechanism for eliminated resources to allow recovery of mistakenly eliminated resources when necessary.

[0358] Metadata update: Update metadata structures such as indexes and caches in a timely manner to ensure that elimination operations do not affect the system's retrieval performance.

[0359] Logging: Record the reasons and process of each elimination operation in detail to facilitate subsequent analysis and auditing.

[0360] Through this multi-dimensional, multi-stage intelligent elimination process, the system can effectively maintain the high quality and efficiency of the resource library while minimizing the risk of misjudgment. This process not only reflects a complex decision-making algorithm, but also reflects a deep understanding of the life cycle and value assessment of educational resources.

[0361] As an exemplary description, optionally, the specific implementation of step 400 includes:

[0362] Based on the resource sorting result obtained in step 300, obtain several resources at the end of the sorting;

[0363] For each of the plurality of resources, the following determination steps are performed:

[0364] Obtain the resource quality index F(x), resource heat index F(y) and resource update index F(z) of the resource;

[0365] Comparing the F(x) with a preset resource quality index threshold x';

[0366] Compare the F(y) with a preset resource heat index threshold y';

[0367] Compare the F(z) with a preset resource update index threshold z';

[0368] The resource is determined to be eliminated if and only if the following conditions are met at the same time:

[0369] F(x)<x'

[0370] F(y) <y‘

[0371] F(z)<z'

[0372] Among them, x', y', z' are preset threshold parameters.

[0373] Step 500: receiving a resource search request from a user, and analyzing the relevance between the user's search statement and each resource in the digital resource library;

[0374] Step 600: Based on the correlation analysis result of step 500 and the resource ranking result of step 300, intelligently recommend resources.

[0375] Optionally, the step 500 specifically includes: analyzing the matching degree between the user search statement and each resource in the digital resource library, the matching degree being selected from the following group: keyword matching degree, synonym matching degree, vocabulary substring or parent string matching degree, core semantic matching degree and resource coding matching degree; and the step 600 specifically includes: calculating the resource recommendation degree based on the matching analysis result of step 500 and the resource sorting result of step 300, and intelligently recommending resources according to the resource recommendation degree.

[0376] Specifically, step 500 and step 600 constitute the intelligent retrieval and recommendation core of the digital teaching resource library, and they work together to provide accurate resource matching and personalized recommendations.

[0377] In step 500, the user's search request is first processed, including query preprocessing (such as word segmentation, spelling correction) and query expansion. Then, the system analyzes the relevance of the user's search statement with the resources in the resource library and adopts a multi-dimensional matching strategy, which exemplarily includes:

[0378] Keyword matching: Use algorithms such as TF-IDF to calculate keyword similarity.

[0379] Synonym matching: Calculate semantic similarity using word embedding technology.

[0380] Word substring or parent string matching: Find the inclusion relationship between words.

[0381] Core semantic matching: Use topic models or deep learning models to understand the overall semantics.

[0382] Resource code matching: Perform specific matching on structured resource codes.

[0383] This multi-dimensional matching ensures the comprehensiveness and accuracy of the retrieval results.

[0384] In step 600, the correlation analysis results and the previous resource ranking results are combined to generate a final recommendation list, which exemplarily includes:

[0385] Feature integration: Convert matching results and sorting information (such as quality index, popularity index, and update index) into feature vectors.

[0386] Recommendation calculation: Use machine learning models (such as random forest or deep learning models) to calculate the comprehensive recommendation.

[0387] Personalized adjustment: Consider user historical behavior and preferences to make personalized recommendations.

[0388] Diversity optimization: Strike a balance between relevance and diversity to avoid results that are too singular.

[0389] Through this process, the system can understand user needs and quickly and accurately recommend the most relevant and valuable educational resources from massive resources, providing a personalized learning experience.

[0390] Optionally, the calculation formula of the resource recommendation degree F2 in step 600 is:

[0391]

[0392] in:

[0393] F′ represents the keyword matching degree between the user's search statement and the resource;

[0394] F″ represents the synonym matching degree between the user’s search statement and the resource;

[0395] F″′ represents the degree of matching between the user’s search statement and the resource’s vocabulary substring or parent string;

[0396] F″″ represents the core semantic matching degree between the user’s search statement and the resource;

[0397] F° represents the matching degree between the user’s search code and the resource code;

[0398] F1 is the resource sorting result obtained in step 300;

[0399] k1 and k2 are weight coefficients, which are used to adjust the importance of matching analysis results and resource ranking results in the recommendation calculation;

[0400] is the weight coefficient, which is used to adjust the importance of semantic matching and encoding matching in the recommendation calculation;

[0401] θ1, θ2, θ3, and θ4 are weight coefficients used to adjust the importance of different types of matching in semantic matching;

[0402] The resource recommendation degree F2 comprehensively considers the matching degree between the user's search requirements and the resources and the comprehensive ranking results of the resources, and is used to determine the priority of recommending resources to the user. The higher the F2 value, the more the resource meets the user's needs and the higher the comprehensive value, and the more priority it is recommended to the user.

[0403] Optionally, the method further comprises:

[0404] Step 700: Feedback the resource elimination information of step (4) to step (1), collect, deduplicate, and classify resources of related topics according to the topics of the eliminated resources to maintain the diversity of resource topics; feed back the resource matching information of step (5) to step (1), collect, deduplicate, and classify resources according to the matching results to enrich the resource types and realize dynamic update and continuous optimization of the digital resource library.

[0405] In order to better understand the technical solution of the present application, a specific example is provided below for illustration. The details listed in the example are mainly for ease of understanding and are not intended to limit the scope of protection of the present application.

[0406] In this example, a method for intelligent management and application of digital teaching resource library is proposed, which specifically includes:

[0407] Step 1: Construction of digital resource library

[0408] The construction of a digital resource library is to use machines to automatically identify the characteristic information, knowledge points and other information of a large amount of collected teaching resources, digitally encode them according to unified coding rules, and deduplicate, classify and store resources based on the resource coding, thereby realizing the automatic construction of a massive digital teaching resource library.

[0409] Specifically, the resource information stored in the resource library includes: resource code, resource management information, resource content feature information, and resource content. Among them, the resource code is a digital unique identifier based on resource attribute information and content feature information, which is used for rapid classification, storage and retrieval of resources; resource management information is the summary information of resource application in various application systems, which is used for resource application behavior data analysis; resource content feature information is other important content feature description information related to resource and teaching application (such as keywords, subject knowledge points, etc.), which is used for resource retrieval and recommendation; resource content is the specific content information of the resource.

[0410] Step 2: Build a resource application big data analysis model

[0411] The resource application big data analysis model is based on the resource application behavior data of each teaching application system. By analyzing the three key indicators of resource quality, resource popularity, and resource update, the result data of resource management is obtained to provide a reference for the intelligent management and application of resource libraries. The specific calculation method of each result data of resource application is as follows:

[0412] (1) Obtaining resource quality index

[0413] In this example, the resource quality index refers to a quantitative indicator used to measure the quality of teaching resources. It can be used to give priority to recommending teaching resources with high quality indexes during intelligent management and application of resource libraries, as well as self-elimination of resources. It can be used to eliminate teaching resources with low quality indexes during intelligent management of resource libraries, and to recommend teaching resources with high quality indexes during intelligent application of resource libraries. The resource quality index is obtained through statistical analysis of resource content applicability, resource content rationality, and user feedback data.

[0414] Specifically, the resource quality index is calculated as follows:

[0415] F(x)=α1*(F(x) (1) )+α2*(F(x) (2) )+α3*(F(x) (3) ) (1)

[0416]

[0417] F(x) (2) =∈1Match{(G1), (R1')}+∈2Match{(G2), (R′2)}+…+∈ m Match{(G m′ ), (R m ')} (3)

[0418]

[0419] In formula (1), F(x) is the quality index of a resource; F(x) (1) is the resource content applicability impact factor, F(x) (2) is the impact factor of resource content rationality, F(x) (3) is the user feedback influencing factor; α1, α2, and α3 are the weight coefficients of each dimension respectively.

[0420] In formula (2), F(x) (1)is the applicability of the resource content; Min(a, b) is the total number of knowledge points covered by the resource, Max(a, b) is the knowledge points that should be mastered in this teaching stage under the teaching objectives; c is the cth subject knowledge point, E c is the mastery requirement of the cth subject knowledge point, and its value is divided into c′, c″, c''' according to the three levels of mastery requirement (i.e., mastery of application, understanding of explanation, and recognition and understanding), and c′>c″>c'''>0; F c is the level of detail of the resource's explanation of c subject knowledge points (the level of detail can be evaluated from the knowledge point's phonetic symbols, word meanings, usage, examples, etc.); Match{(E c ), (F c )} is the matching result between the mastery requirement of the cth subject knowledge point and the detail level of the explanation content of the cth subject knowledge point in the resource; d is the dth general knowledge point, E d is the mastery requirement of the dth general knowledge point, and its value is divided into d′, d″, d''' according to the three levels of mastery requirement, and d′>d″>d'''>0; F d is the level of detail of the resource's explanation of d general knowledge points; Match{(E d ), (F d )) is the matching result between the mastering requirement of the d-th general knowledge point and the level of detail of the resource’s explanation of the d-th general knowledge point; ε1, ε2, and ε3 are weight coefficients.

[0421] In formula (3), F(x) (2) The rationality of resource content; G1, G2, G m are the mastery requirements of the 1st, 2nd, and mth knowledge points contained in the resource at this teaching stage. The values ​​are divided into g′, g″, and g''' according to the three levels of mastery requirements (i.e., mastering application, understanding explanation, and knowing understanding), and g′>g″>g'''>0; R1, R′2, R m are the average cognitive levels of the teaching group for the 1st, 2nd, and mth knowledge points in this teaching stage; Match{(G1), (R1), Match{(G2), (R′2)}, Match{(G m′ ), (R m ) are the matching results between the mastering requirements of the 1st, 2nd, and mth knowledge points of the resource at this teaching stage and the average cognitive level of the teaching group on the 1st, 2nd, and mth knowledge points; ∈1, ∈2, ∈ m is the weight coefficient.

[0422] In formula (4), F(x) (3) is the user feedback result; N is the total number of resources contained in the sub-library to which the resource belongs, K n The total number of times the nth resource is used by teachers for teaching, K(应用该资源) is the total number of times the resource is used by teachers for teaching, K (反馈该资源) The total number of times teachers give feedback that the resource is a high-quality resource after using it for teaching; L n The total number of times the nth resource is used by students for learning, L (应用该资源) is the total number of times the resource is used by students for learning, L (反馈该资源) It is the total number of times that students feedback that the resource is a high-quality resource after applying it for learning; μ1, μ2, μ3, and μ4 are weight coefficients.

[0423] (2) Obtaining resource heat index

[0424] The resource heat index is a quantitative indicator used to measure the popularity or attention of a resource. It can be used to give priority to recommending teaching resources with a high heat index and self-elimination of resources during intelligent management and application of resource libraries. The resource heat index is obtained through statistical analysis of resource application trend, resource application frequency, and user application feedback data.

[0425] Specifically, the resource heat index is calculated as follows:

[0426] F(y)=β1*(F(y) (1) )+β2*(F(y) (2) )+β3*(F(y) (3) ) (5)

[0427]

[0428]

[0429]

[0430] In formula (5), F(y) is the heat index of a resource; (1) is the impact factor of resource application trend, F(y) (2) is the resource application frequency influence factor, F(y) (3) is the user application feedback influencing factor; β1, β2, and β3 are the weight coefficients of each dimension.

[0431] In formula (6), F(y) (1) T is the resource application trend degree; 分析 is the interval between two resource heat analyses, u′ is the total number of times the resource is used by teachers for teaching on the kth day, and d ku is the teacher's application behavior for the resource on the kth day uth time, s 1u is the weight value of the teacher's application behavior of the resource for the uth time; v′ is the total number of times the resource is applied by students for learning on the kth day, e kvis the student's application behavior for the resource on the kth day and the vth time, s 2v is the weight value of the student's application behavior for the vth time of this resource; It refers to the date weight value of the kth day; π1 and π1 are weight coefficients.

[0432] Note: ① When users with different identities use resources, their application behaviors are different due to their identity permissions. For example, for a resource with student identity, its application behaviors are only browsing and downloading, while for teacher identity, its application behaviors are browsing, downloading, selecting, editing, and abandoning. ② The weight values ​​of resources in different application behaviors are different. For example, the browsing behavior index is recorded as 1, the downloading behavior index is recorded as 2, the editing behavior index is recorded as -1, the selecting behavior index is recorded as 5, and the abandoning behavior index is recorded as -2. ③ Date weight The standard is: the weight of the date applied within this week is recorded as 4, the weight of the date applied in the previous week is recorded as 3, and the weight of the date applied in the previous two weeks is recorded as 2....

[0433] In other words, in the application of digital teaching resource libraries, there are significant differences in the application behaviors of users with different identities, which is mainly due to the different permissions they have. For example, for student users, their resource application behaviors are usually limited to browsing and downloading. This reflects the main needs of students as learners, that is, to obtain and use learning materials. In contrast, teacher users have a wider range of application behaviors, including browsing, downloading, selecting, editing, and abandoning. This difference reflects the multiple roles of teachers in the teaching process, who not only need to use resources, but also need to select, modify, and manage resources.

[0434] In order to more accurately evaluate the impact of different application behaviors on the value of resources, the system assigns different weight values ​​to various behaviors. This weight setting reflects the different contributions of various behaviors to the importance and popularity of resources. For example, browsing behavior is given a relatively low weight, such as 1, indicating that this is a basic and common way of use. The weight of downloading behavior is slightly higher, such as 2, because it implies a higher user interest in the resource. The selection behavior is given a higher positive weight, such as 5, which indicates that the resource is actually used in teaching and has a high value.

[0435] In addition, the system also takes into account the time factor of resource usage, and reflects the timeliness of resources through date weights. Recently used resources are given higher weights, reflecting their current relevance and popularity. Specifically, resources applied within this week receive the highest date weight of 4, resources applied in the previous week receive a weight of 3, and the weights of resources before that gradually decrease. This design ensures that the system can prioritize the latest and most relevant resources, while not completely ignoring valuable resources that have been used earlier.

[0436] Through this sophisticated weight design, the system can comprehensively and dynamically evaluate the actual value and popularity of each resource, thus providing a solid data foundation for intelligent resource management and recommendation. This not only improves the efficiency of resource utilization, but also better meets the diverse needs of different user groups.

[0437] In formula (7), F(y) (2) is the frequency of resource application; i′ is the total number of teachers who use all resources, h′ is the total number of times the resource is used by teachers in teaching, and b ih The resource is used by the i-th teacher for the hth time, t ih is the time when the resource is used by the i-th teacher for the hth time; j′ is the total number of students who use all resources, g′ is the total number of times the resource is used by students for learning, c jg The resource is used by the jth student for the gth time, t jg is the application time of the resource by the j-th student for the g-th time; ρ1 and ρ2 are weight coefficients.

[0438] In formula (8), F(y) (3) Provide feedback results to users; s is the total number of times the resource is used by teachers for teaching, P′ s The number of effective feedback behaviors (likes, favorites, comments, etc.) after the resource is used by teachers for teaching; Q r is the total number of times the resource is used by students for learning, Q′ r It is the number of effective feedback behaviors (likes, favorites, comments, etc.) after students apply and learn the resource; σ1 and σ1 are weight coefficients.

[0439] (3) Obtain resource update index

[0440] The resource update index is a quantitative indicator used to measure the degree of resource update. It can be used to give priority to recommending teaching resources with high update indexes and self-elimination of resources during intelligent management and application of resource libraries. The resource update index is obtained through statistical analysis of the update time, update degree, update effect and other aspects of various resources.

[0441] Specifically, the resource update index is calculated as follows:

[0442] F(z)=γ1*(F(z) (1) )+γ2*(F(z) (2) )+γ3*(F(z) (3) ) (9)

[0443]

[0444]

[0445]

[0446] In formula (9), F(z) is the update index of a resource; (1) is the resource update time impact factor, F(z) (2) is the resource renewal degree influencing factor, F(z) (3) is the influencing factor of the user feedback update effect; γ1, γ2, and γ3 are the weight coefficients of each dimension respectively.

[0447] In formula (10), F(z) (1) T is the resource update time; 统计2 It refers to the time when the system is currently performing resource analysis. 统计1 Refers to the time when the system last performed resource analysis, T 入库 Refers to the storage time of the resource; T 更新 It refers to the latest update time of the resource; ω1 and ω2 are weight coefficients.

[0448] In formula (11), F(z) (2) is the resource update degree; T 统计2 It refers to the time when the system is currently performing resource analysis. 统计1 Refers to the time when the system last performed resource analysis. 更新 It refers to the number of times the resource is updated between two resource analysis intervals; S′ 更新 It refers to the sum of data updates within the time interval between two resource analyses, S′ 原始 refers to the total amount of original data, and w refers to the adjustment coefficient; is the weight coefficient.

[0449] In formula (12), F(z) (3) Provide user feedback on update effects; R′ p is the number of times the teacher gives feedback on the resource update effect after applying it for teaching, p′ is the total number of times the teacher gives feedback on the resource update effect, R p is the teacher's feedback result on the resource data update for the pth time; W′ q is the number of times the resource is updated significantly after students apply it for learning, q′ is the total number of times students provide feedback on the resource update effect, and W q is the feedback result of the student's qth update of the resource data; τ1 and τ2 are weight coefficients.

[0450] Step 3: Intelligent management and application of digital resource library

[0451] This part realizes the intelligent management and application of digital resource libraries based on the construction of massive digital resource libraries and the result data of big data analysis of resource applications. Specifically, in terms of resource library management, the digital resource library is refined and streamlined through intelligent sorting and elimination of resources; in terms of resource library application, intelligent matching and recommendation generation of refined resources that meet user search needs are achieved through intelligent recommendation of resources.

[0452] (1) Intelligent sorting and elimination of resources to achieve intelligent management of digital resource libraries

[0453] By combining the big data analysis results of resource applications such as the resource recommendation index, resource update index, and resource popularity index, resources are intelligently sorted and eliminated, so that the resources in the resource library are intelligently arranged from high to low according to the sorting results, so as to retain the high-quality resources at the top of the sort and eliminate the low-quality resources at the end of the sort, thus realizing the intelligent management of resources in the resource library.

[0454] ① Intelligent sorting of resources: Specifically, the calculation method of resource sorting is as follows:

[0455] F1=δ1*F(x)+δ2*F(y)+δ3*F(z)(13)

[0456] In formula (13), F1 represents the ranking result of each resource; F(x) is the quality index of each resource, F(y) is the popularity index of each resource, and F(z) is the update index of each resource; δ1, δ2, and δ3 are the weight coefficients of each dimension respectively.

[0457] ② Intelligent resource elimination: Based on the resource sorting results, obtain several resources at the end of the sorting, and then obtain the resource quality index, resource heat index, and resource update index of these resources respectively, and further determine: the resource quality index F(x) reference threshold x', the resource heat index F(y) reference threshold y', and the resource update index F(z) reference threshold z'. If the resource's F(x) is less than x', F(y) is less than y', and F(z) is less than z', then the resource is determined to be eliminated.

[0458] Note: ① The advantage of resource sorting is that it not only retains high-quality, popular, and new resources, but also regularly eliminates poor-quality, low-popular, and old resources, ensuring that the resource library is full of high-quality resources while saving storage space for the resource library; ② It allows the elimination information to be fed back to the resource construction module, and the resource construction module collects resources of related topics, removes duplications, and classifies them into the database based on the topics of the eliminated resources, thereby maintaining the diversity of resource topics.

[0459] In other words, in the intelligent management system of the digital teaching resource library, the resource sorting and elimination mechanism plays a key role, providing strong support for the efficient operation and continuous optimization of the resource library. The design concept of this mechanism not only reflects the strict control of resource quality, but also reflects the system's deep understanding of the dynamic management of the resource library.

[0460] First, the core advantage of the resource ranking mechanism is that it can comprehensively evaluate multiple key dimensions of resources. By comprehensively considering the quality, popularity, and novelty of resources, the system can accurately identify and highlight the most valuable resources. This approach ensures that high-quality, popular, and updated resources are given priority display and recommendation, thereby improving the efficiency of the entire resource library and user satisfaction. At the same time, this ranking method also provides a scientific basis for resource elimination.

[0461] Resource elimination is a natural extension of the resource sorting mechanism. By regularly identifying and removing resources that are of poor quality, have low usage rates, and have not been updated for a long time, the system effectively maintains the overall quality of the resource library. This "survival of the fittest" strategy not only ensures that the resource library always maintains high-quality resources, but also brings significant storage space savings. With the rapid growth of digital resources today, the importance of this space optimization strategy is self-evident. It enables limited storage resources to be used more efficiently and creates conditions for the continuous introduction of new high-quality resources.

[0462] However, resource elimination does not mean simply deleting resources. Another innovation of the system is that it establishes a feedback loop mechanism. When a resource is eliminated, this information will be passed back to the resource construction module. This module will then carry out targeted operations such as collecting, deduplicating, and classifying new resources into the database based on the subject characteristics of the eliminated resources. This mechanism cleverly maintains the subject diversity of the resource library and ensures that even if some resources are eliminated, there will be no gaps in the knowledge coverage of related topics.

[0463] This dynamic balance is achieved by taking into account the forward-looking and comprehensive nature of the system design. It not only focuses on optimizing the current state of the resource library, but also takes into account long-term continuous improvement. In this way, the resource library can continue to evolve, adapt to the ever-changing educational needs and knowledge structure, and always maintain its status as a collection of high-quality teaching resources.

[0464] In general, this resource sorting and elimination mechanism, combined with the feedback loop system, creates a self-improving, dynamically optimized intelligent resource management ecosystem. It not only improves the overall quality and practicality of resources, but also ensures the comprehensiveness and up-to-dateness of resource topics, truly realizing the efficient management and continuous optimization of digital teaching resources.

[0465] (2) Intelligent resource recommendation to realize intelligent application of digital resource library

[0466] By analyzing the matching degree between the user's search statement (or search code) and each resource in the resource library, combined with the resource sorting results, intelligent matching and recommendation are generated with high matching degree and quality resources of the search statement, realizing the intelligent application of the resource library. The calculation method of resource recommendation is as follows:

[0467]

[0468] In formula (14), F2 represents the recommendation result of the resource; F′ represents that the sentence retrieved by the user contains the same keyword as each resource, F″ represents that the sentence retrieved by the user contains the synonyms (synonyms) of the same keyword as each resource, F″′ represents that the sentence retrieved by the user contains the substring or parent string of related words as each resource, and F″″ represents that the sentence retrieved by the user contains the same core semantics as each resource; F° represents the matching result of the user retrieval code and the code of each resource; F1 represents the ranking result of each resource, k1, k2, θ1, θ2, θ3, and θ4 are weight coefficients respectively.

[0469] Note: ① The benefit of the resource recommendation design is that it not only performs matching analysis between the user's search statement and the resources in the resource library, such as keywords, synonyms, semantics, etc., to improve the accuracy of the search; it also considers the ranking results of the resources, and gives priority to recommending the highest quality, most popular, and latest high-quality resources to users, which is conducive to users' targeted teaching and learning. ② It allows the matching information to be fed back to the resource construction module, and the resource construction module collects, removes duplicates, and classifies and stores the search resources according to the matching results (if the search resources are not matched), enriching the types of resources.

[0470] In other words, the resource recommendation mechanism of the digital teaching resource library is a carefully designed multi-dimensional intelligent system, whose core goal is to provide users with the most relevant and highest quality teaching resources. The design concept of this mechanism not only reflects a deep understanding of user needs, but also demonstrates the system's innovative thinking in resource matching and quality control.

[0471] First, the resource recommendation system adopts a multi-level matching analysis strategy. When a user enters a search statement, the system not only performs a simple keyword match, but also uses multiple indicators including keywords, synonyms, and deeper semantic analysis for comprehensive matching. This multi-dimensional matching method greatly improves the accuracy and comprehensiveness of the search. For example, even if the terms used by the user are not exactly the same as the description in the resource, the system can find relevant resources through synonyms and semantic analysis. This method effectively solves the limitations that traditional keyword searches may bring, making it easier for users to find resources that truly meet their needs, even if they may not know exactly what keywords to use.

[0472] Secondly, the resource recommendation mechanism also cleverly integrates the results of resource sorting. This means that on the basis of matching relevance, the system also considers factors such as resource quality, popularity, and update time. In this way, the recommendation system can not only find resources relevant to user needs, but also ensure that the recommended resources are the highest quality, most popular, and latest. This design concept reflects the system's comprehensive understanding of "relevance" - truly relevant resources should not only match user needs in terms of content, but also be of high quality and timeliness. This comprehensive consideration greatly enhances the practical value of recommended resources, allowing users to quickly obtain the best quality and most suitable teaching or learning materials, thereby improving the efficiency and effectiveness of teaching and learning.

[0473] Furthermore, the innovation of this recommendation system lies in its self-improvement mechanism. The system allows matching information to be fed back to the resource construction module, which creates a dynamic optimization cycle. When the system finds that some of the user's search requirements cannot be met (for example, no fully matching resources are found), this information will be passed to the resource construction module. The resource construction module will then carry out targeted operations such as collecting, deduplicating, and classifying new resources into the database based on these unmet search requirements. This mechanism ensures that the resource library can be continuously expanded and updated to meet the ever-changing needs of users.

[0474] This adaptive resource expansion strategy brings multiple benefits. First, it ensures that the resource library keeps pace with the times and can respond to emerging educational topics and needs in a timely manner. Second, it improves the comprehensiveness and diversity of the resource library, allowing the resource library to cover a wider range of educational topics and resource types. Finally, this dynamic adjustment mechanism can also identify and fill knowledge gaps in the resource library, ensuring that the resource library can fully meet the diverse needs of various users.

[0475] In general, this multi-dimensional, intelligent resource recommendation mechanism, combined with a dynamic resource expansion strategy, creates an evolving, self-improving educational resource ecosystem. It not only greatly improves the efficiency and accuracy of users' access to relevant resources, but also ensures the continuous optimization and expansion of the resource library content. This design fully reflects the cutting-edge concept of digital teaching resource management and lays a solid foundation for providing personalized, high-quality educational resource support.

[0476] The above embodiments have the following technical effects:

[0477] The above embodiments have brought significant technical effects in the field of educational resource management and application. First, it realizes the efficient management and precise application of teaching resources. Through the comprehensive evaluation of resource quality index, heat index and update index, the method conducts a comprehensive quantitative evaluation of resources. Based on these evaluation results, the system can perform intelligent sorting and elimination, effectively ensuring the high quality of resources in the resource library. At the same time, the intelligent matching and recommendation functions greatly improve the accuracy and efficiency of resource retrieval, allowing educators and learners to find the required high-quality resources more quickly.

[0478] Secondly, the above embodiment significantly improves the dynamic optimization capability of the resource library. The resource application big data analysis model realizes real-time monitoring of resource usage, and the resource library can be continuously updated and optimized through the feedback mechanism of resource elimination and matching information. This mechanism not only ensures the timeliness of the content of the resource library, but also maintains the diversity of resource themes, so that the resource library can always meet the latest needs of teaching and learning.

[0479] In terms of teaching effect, the application of the above embodiments improves the utilization rate of teaching resources and the teaching quality. The intelligent recommendation of high-quality resources helps teachers choose the most suitable teaching materials, thereby improving the teaching quality. The precise matching of resources meets the personalized needs of different users. Whether it is teachers preparing lessons or students learning independently, they can find the most suitable resources.

[0480] From a management perspective, the above embodiment saves a lot of storage space and management costs. The intelligent elimination mechanism reduces the occupation of redundant and obsolete resources, and the automated management process greatly reduces the cost of manual maintenance. The efficient use of resources also improves the performance of the overall system, allowing limited hardware resources to support more users' simultaneous access and use.

[0481] Finally, the above embodiments provide important data support for educational decision-making. By analyzing resource application data, educational administrators and teachers can obtain valuable evidence for improving teaching content and methods. User feedback data helps to understand learners' needs and learning effects, thereby formulating more targeted teaching strategies and learning plans.

[0482] In summary, the above embodiments achieve efficient management and accurate application of teaching resources through intelligent means, providing strong support for improving teaching quality and learning outcomes.

[0483] The second embodiment of the present application relates to a digital teaching resource library intelligent management and application system, the structure of which is as follows: Figure 3 As shown, the digital teaching resource library intelligent management and application system includes:

[0484] A digital resource library construction module is used to collect teaching resources, digitally process and store the teaching resources, and construct a digital resource library;

[0485] The resource application big data analysis module is used to construct a resource application big data analysis model based on the resource application behavior data in the digital resource library, and calculate the following indicators for subsequent resource management and application: a) resource quality index, including resource content applicability, resource content rationality and user feedback data; b) resource popularity index, including resource application trend degree, resource application frequency degree and user application feedback data; c) resource update index, including resource update time, update degree and update effect;

[0486] A resource intelligent sorting module, used to intelligently sort the resources in the digital resource library based on the resource quality index, resource heat index and resource update index obtained by the resource application big data analysis module to obtain a resource sorting result;

[0487] A resource intelligent elimination module is used to intelligently eliminate resources at the end of the sorting and whose three indexes are all lower than the corresponding thresholds according to the resource sorting results of the resource intelligent sorting module and in combination with a preset resource quality index threshold, a resource heat index threshold and a resource update index threshold;

[0488] A resource search analysis module, used to receive a user's resource search request and analyze the relevance between the user's search statement and each resource in the digital resource library;

[0489] The resource intelligent recommendation module is used to make intelligent recommendations on resources based on the correlation analysis results of the resource retrieval analysis module and the resource sorting results of the resource intelligent sorting module.

[0490] The first implementation manner is a method implementation manner corresponding to the present implementation manner. The technical details in the first implementation manner can be applied to the present implementation manner, and the technical details in the present implementation manner can also be applied to the first implementation manner.

[0491] It should be noted that those skilled in the art should understand that the implementation functions of each module shown in the above-mentioned implementation of the digital teaching resource library intelligent management and application system can be understood by referring to the relevant description of the above-mentioned digital teaching resource library intelligent management and application method. The functions of each module shown in the implementation of the above-mentioned digital teaching resource library intelligent management and application system can be realized by a program (executable instruction) running on a processor, or by a specific logic circuit. If the above-mentioned digital teaching resource library intelligent management and application system of the present application embodiment is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, Read Only Memory), a disk or an optical disk. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.

[0492] Accordingly, an embodiment of the present application further provides a computer storage medium, in which computer executable instructions are stored. When the computer executable instructions are executed by a processor, the various method embodiments of the present application are implemented.

[0493] All documents mentioned in this application are considered to be included in the disclosure of this application as a whole, so that they can be used as the basis for modification when necessary. In addition, it should be understood that after reading the above disclosure of this application, those skilled in the art can make various changes or modifications to this application, and these equivalent forms also fall within the scope of protection claimed in this application.

Claims

1. A method for intelligent management and application of digital teaching resource library, characterized in that: The following steps are involved: Step 100: constructing a digital resource library, including collecting teaching resources, and digitally processing and storing the teaching resources; Step 200: construct a resource application big data analysis model, which is based on the resource application behavior data in the digital resource library and calculates the following indicators for subsequent resource management and application: a) resource quality index, including resource content applicability, resource content rationality and user feedback data; b) resource popularity index, including resource application trend degree, resource application frequency degree and user application feedback data; c) resource update index, including resource update time, update degree and update effect; Step 300: Based on the resource quality index, resource heat index and resource update index obtained in step 200, intelligently sort the resources in the digital resource library to obtain a resource sorting result; Step 400: According to the resource sorting result of step 300, in combination with the preset resource quality index threshold, resource heat index threshold and resource update index threshold, the resources at the end of the sorting and whose three indexes are all lower than the corresponding thresholds are intelligently eliminated; Step 500: receiving a resource search request from a user, and analyzing the relevance between the user's search statement and each resource in the digital resource library; Step 600: Based on the correlation analysis result of step 500 and the resource ranking result of step 300, intelligently recommend resources.

2. The method according to claim 1, characterized in that: The calculation formula of the resource quality index F(x) is: F(x)=α1*(F(x) (1) )+α2*(F(x) (2) )+α3*(F(x) (3) ) (1) in: F(x) (1) Indicates the applicability of resource content, reflecting the degree of match between resource content and teaching objectives, including the number of knowledge points covered by the resource, the matching between the requirements for mastering knowledge points and the level of detail of resource content; F(x) (2) Indicates the rationality of resource content, reflecting the matching degree between resource content and the teaching stage and the cognitive level of the teaching group, including the matching between the mastery requirements of each knowledge point in the resource and the average cognitive level of the corresponding knowledge point of the teaching group at this teaching stage; F(x) (3) It represents the user feedback results, reflecting the evaluation of teachers and students on the quality of resources, including the frequency of resource application and the proportion of resources that are fed back as high-quality resources; α1, α2, and α3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource quality assessment.

3. The method according to claim 2, characterized in that The calculation of the resource quality index F(x) also includes the following specific formula: in: For formula (2): Min(a,b) represents the total number of knowledge points covered by the resource Max(a,b) represents the total number of knowledge points that should be mastered in this teaching stage under the teaching objectives E c Indicates the requirements for mastering the cth subject knowledge point F c Indicates the level of detail of the resource's explanation of the cth topic knowledge point E d Indicates the mastery requirement of the dth general knowledge point F d Indicates the level of detail of the resource's explanation of the dth general knowledge point The Match{} function indicates the degree of match between two parameters. ε1, ε2, ε3 are weight coefficients For formula (3): G k (k=1,2,...,m) represents the mastery requirement of the kth knowledge point contained in the resource at this teaching stage R k '(k=1,2,...,m) represents the average cognitive level of the teaching group on the kth knowledge point in this teaching stage ∈ k (k=1,2,...,m) is the weight coefficient For formula (4): K (apply this resource) indicates the total number of times this resource is used by teachers in teaching. K (Feedback on this resource) indicates the total number of times teachers give feedback that this resource is a high-quality resource after using it for teaching. L (应用该资源) Indicates the total number of times the resource has been used by students for learning L (反馈该资源) Indicates the total number of times students give feedback that the resource is a high-quality resource after applying it for learning. K (n) and L n Represents the total number of times the nth resource is used by teachers and students respectively N represents the total number of resources contained in the sub-library to which the resource belongs μ1, μ2, μ3, μ4 are weight coefficients These formulas together constitute the detailed calculation method of the resource quality index F(x), which comprehensively evaluates the content applicability, content rationality and user feedback of the resource.

4. The method according to claim 1, characterized in that The calculation formula of the resource heat index F(y) is: F(y)=β1*(F(y) (1) )+β2*(F(y) (2) )+β3*(F(y) (3) ) (5) in: F(y) (1) It indicates the degree of resource application trend, reflecting the application trend of resources in a certain period of time, including the number, frequency and time distribution of teachers' application teaching and students' application learning; F(y) (2) Indicates the frequency of resource application, reflecting the overall situation of resource use by teachers and students, including the total number of times the resource is used, the total duration, and the proportion of its application in all resources; F(y) (3) It indicates the user application feedback results, reflecting the direct feedback of teachers and students after applying the resources, including the proportion of effective feedback behaviors such as likes, collections, and comments; β1, β2, and β3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource heat assessment.

5. The method according to claim 4, characterized in that The calculation of the resource heat index F(y) also includes the following specific formula: in: For formula (6): T 分析 Indicates the interval between two resource heat analyses d ku Indicates the teacher's application behavior of the resource on the kth day uth time e kv Indicates the student's application behavior for the resource on the kth day and the vth time s 1u Indicates the weight value of the teacher's application behavior for the uth time of this resource s 2v Indicates the weight value of the v-th student application behavior of this resource Indicates the date weight value of the kth day u′ represents the total number of times the resource is used by teachers for teaching on the kth day v′ represents the total number of times the resource is used by students for learning on the kth day π1 is the weight coefficient For formula (7): b ih Indicates that the resource is ih Teachers h times applied t ih Indicates that the resource is ih The application time of the hth application by each teacher c jg Indicates that the resource is used by the jth student for the gth time t jg Indicates the time when the resource was used by the jth student for the gth time i′ represents the total number of teachers who apply all resources h′ represents the total number of times the resource is used by teachers in teaching j′ represents the total number of students who applied all resources g′ represents the total number of times the resource is used by students for learning ρ1, ρ2 are weight coefficients For formula (8): P s Indicates the total number of times the resource has been used by teachers for teaching P s ′ indicates the number of times the teacher provides effective feedback after using the resource in teaching Q r Indicates the total number of times the resource has been used by students for learning Q r ′ represents the number of times students provide effective feedback after applying the resource σ1, σ1 are weight coefficients These formulas together constitute the detailed calculation method of the resource heat index F(y), which comprehensively evaluates the application trend, application frequency and user feedback of the resource.

6. The method according to claim 1, characterized in that The calculation formula of the resource update index F(z) is: F(z)=γ1*(F(z) (1) )+γ2*(F(z) (2) )+γ3*(F(z) (3) ) (9) in: F(z) (1) Indicates resource update time, reflecting the timeliness of resources, including the relationship between resource storage time, latest update time and current time; F(z) (2) Indicates the resource update degree, reflecting the amount of change in resource content, including the number of resource updates within a certain time interval, and the ratio of the updated data volume to the original data volume; F(z) (3) Indicates the effect of user feedback on the update, reflecting the evaluation of teachers and students on the effect of the resource update, including the proportion of updated resources that were fed back as having significant effects; γ1, γ2, and γ3 are weight coefficients of each dimension, which are used to adjust the importance of different factors in resource renewal assessment; Among them, ε1, ε2, ε3, β1, β2, β3, γ1, γ2, and γ3 are all weight coefficients, which are used to adjust the importance of each factor in the calculation of resource quality index, resource heat index, and resource update index. These weight coefficients can be adjusted according to specific application scenarios and needs.

7. The method according to claim 6, characterized in that The calculation of the resource update index F(z) also includes the following specific formula: in: For formula (10): T 统计2 Indicates the time when the system is currently performing resource analysis T 统计1 Indicates the time when the system last performed resource analysis T 入库 Indicates the time when the resource was put into storage T 更新 Indicates the latest update time of the resource ω1, ω2 are weight coefficients For formula (11): S 更新 Indicates the number of times the resource is updated between two resource analysis intervals S′ 更新 Indicates the sum of data updates between two resource analysis intervals S′ 原始 Indicates the total amount of raw data w represents the adjustment factor is the weight coefficient For formula (12): R′ p Indicates the number of times the resource has been updated with significant feedback from teachers after being used in teaching R p Indicates the teacher's feedback result on the resource data update for the pth time p′ represents the total number of times teachers provide feedback on the effectiveness of resource updates W′ q Indicates the number of times the resource has been updated with significant feedback from students after application and learning. W q Indicates the feedback result of the student's qth update of the resource data q′ represents the total number of times students provide feedback on the effect of resource updating τ1, τ2 are weight coefficients These formulas together constitute the detailed calculation method of the resource update index F(z), which comprehensively evaluates the update time, update degree and update effect of the resource. Formula (10) reflects the update time of the resource, formula (11) measures the update degree of the resource, and formula (12) evaluates the user's feedback on the resource update effect.

8. The method according to claim 1, characterized in that The calculation formula of the resource ranking result z1 in step 300 is: F1=δ1*F(x)+δ2*F(y)+δ3*F(z) (13) in: F(x) is the resource quality index calculated in step 200, reflecting the content quality of the resource; F(y) is the resource heat index calculated in step 200, reflecting the popularity of the resource; F(z) is the resource update index calculated in step 200, reflecting the timeliness of the resource; δ1, δ2, and δ3 are weight coefficients used to adjust the importance of resource quality, popularity, and update in resource sorting. These weight coefficients can be adjusted according to specific application scenarios and needs; The resource ranking result F1 comprehensively considers the quality, popularity and timeliness of the resources, and is used to comprehensively evaluate and sort the resources in the digital resource library. The higher the F1 value, the higher the comprehensive value of the resource, and the higher the ranking in the resource library.

9. The method according to claim 1, characterized in that The step 500 specifically includes: analyzing the matching degree between the user search statement and each resource in the digital resource library, wherein the matching degree is selected from the following group: keyword matching degree, synonym matching degree, vocabulary substring or parent string matching degree, core semantic matching degree and resource encoding matching degree; and The step 600 specifically includes: calculating the resource recommendation degree based on the matching degree analysis result of step 500 and the resource ranking result of step 300, and intelligently recommending resources according to the resource recommendation degree; and, The calculation formula of the resource recommendation degree F2 in step 600 is: in: F′ represents the keyword matching degree between the user's search statement and the resource; F″ represents the synonym matching degree between the user’s search statement and the resource; F″′ represents the degree of matching between the user’s search statement and the resource’s vocabulary substring or parent string; F″″ represents the core semantic matching degree between the user’s search statement and the resource; F° represents the matching degree between the user’s search code and the resource code; F1 is the resource sorting result obtained in step 300; k1 and k2 are weight coefficients, which are used to adjust the importance of matching analysis results and resource ranking results in the recommendation calculation; is the weight coefficient, which is used to adjust the importance of semantic matching and encoding matching in the recommendation calculation; θ1, θ2, θ3, and θ4 are weight coefficients used to adjust the importance of different types of matching in semantic matching; The resource recommendation degree F2 comprehensively considers the matching degree between the user's search requirements and the resources and the comprehensive ranking results of the resources, and is used to determine the priority of recommending resources to the user. The higher the F2 value, the more the resource meets the user's needs and the higher the comprehensive value, and the higher the priority recommended to the user.

10. A digital teaching resource library intelligent management and application system, characterized in that: include: A digital resource library construction module is used to collect teaching resources, digitally process and store the teaching resources, and construct a digital resource library; The resource application big data analysis module is used to construct a resource application big data analysis model based on the resource application behavior data in the digital resource library, and calculate the following indicators for subsequent resource management and application: a) resource quality index, including resource content applicability, resource content rationality and user feedback data; b) resource popularity index, including resource application trend degree, resource application frequency degree and user application feedback data; c) resource update index, including resource update time, update degree and update effect; A resource intelligent sorting module, used to intelligently sort the resources in the digital resource library based on the resource quality index, resource heat index and resource update index obtained by the resource application big data analysis module to obtain a resource sorting result; A resource intelligent elimination module is used to intelligently eliminate resources at the end of the sorting and whose three indexes are all lower than the corresponding thresholds according to the resource sorting results of the resource intelligent sorting module and in combination with a preset resource quality index threshold, a resource heat index threshold and a resource update index threshold; A resource search analysis module, used to receive a user's resource search request and analyze the relevance between the user's search statement and each resource in the digital resource library; The resource intelligent recommendation module is used to make intelligent recommendations on resources based on the correlation analysis results of the resource retrieval analysis module and the resource sorting results of the resource intelligent sorting module.

Images