Digital teaching resource calling management system and method based on online teaching

Abstract

The invention provides a digital teaching resource calling management system and method based on online teaching, and relates to the technical field of resource library management, and the system comprises a resource collection and construction module which is used for collecting data and constructing a teaching resource model, and an authority management module which is used for extracting a user type according to a Gaussian mixture algorithm, and creating a login channel through the user type; the user permission module is used for performing permission endowing on each user type according to a login channel to obtain user permission data, updating the user permission data according to user activities and historical search data, and the model construction and application module is used for constructing a personalized learning model and optimizing the search performance of the personalized learning model by using the historical search data. And inputting user learning data to obtain customized learning planning resource data, wherein the feedback driving optimization module is used for optimizing the teaching resource model according to feedback data. According to the invention, by adjusting the user permission and establishing the personalized learning model, the learning plan is customized for the user, resource calling and recommendation are optimized, and the utilization rate and accuracy of teaching resources are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of resource library management, and in particular to a digital teaching resource retrieval management system and method based on online teaching. Background Art

[0002] The prototype of online teaching is to provide learning materials through email and basic websites. Early online courses were mostly recorded videos and simple text materials with weak interactivity. With the development of the Internet, learning management systems (such as Blackboard and Moodle) emerged, providing educational institutions with tools to manage courses and students. The concept of massive open online courses (MOOCs) and the rise of platforms such as Coursera and edX marked the beginning of a new era for online teaching. These platforms offer a variety of courses, attracting learners around the world and promoting the concept of open education. The popularity of smartphones has promoted the development of mobile learning, allowing learners to access learning resources anytime and anywhere. Education technology companies have begun to use data analysis and artificial intelligence to provide personalized learning experiences, recommending resources and courses based on users' learning paths and preferences. Educational institutions use online platforms such as Zoom and Google Classroom for teaching, and many traditional classroom teaching methods are forced to adapt to the online environment. Faced with the sharp increase in online learning demand, institutions have begun to pay attention to the quality of online teaching, adopt more systematic evaluation standards, and promote improvements in teacher training and course design.

[0003] The digital teaching resource retrieval management system is an integrated software platform or system designed to provide comprehensive support for online education and blended learning environments. It manages and distributes various forms of digital teaching materials, such as video lectures, e-books, exercises, exams, and interactive learning tools, in an automated and intelligent way, and allows teachers and students to easily access these resources.

[0004] In the prior art, the recommendation mechanism of the teaching resource retrieval management system is single and cannot meet the personalized learning needs of different users. At the same time, it is difficult to adjust user permissions. When the user type changes, the user permissions cannot be changed quickly and inaccurate searches are encountered during the retrieval process, resulting in a mismatch between actual needs and search content. Summary of the invention

[0005] The present invention provides a digital teaching resource retrieval management system based on online teaching, which is used to solve the defects of the prior art that a unified recommendation mechanism is adopted, resulting in failure to meet the personalized learning needs of different users and failure to adjust permissions in real time according to user needs, and lack of accuracy when searching for teaching resources.

[0006] On the one hand, the present invention provides a digital teaching resource retrieval management system based on online teaching, comprising: Resource collection and construction module: used to collect user learning data, teaching resources and historical search data in real time, classify teaching resources according to preset categories to obtain resource statistics, and build a teaching resource model based on resource statistics.

[0007] Permission management module: used to extract user types from user learning data according to the Gaussian mixture model algorithm at a preset period, create a login channel according to the user type, grant permissions to each user type according to the login channel to obtain user permission data, and update the user permission data based on user activities and historical search data.

[0008] Model building and application module: build a personalized learning model based on the teaching resource model, user authority data and historical search data, input the historical search data into the personalized learning model for training, optimize the search performance of the personalized learning model, input the user learning data into the personalized learning model, and obtain customized learning planning resource data.

[0009] Feedback-driven optimization module: set up feedback optimization channels in the teaching resource model, obtain feedback data after users use the personalized learning model, and optimize the teaching resource model in real time based on the feedback data.

[0010] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the authority management module includes: User data analysis unit: used to pre-process user learning data, analyze the pre-processed user learning data through the Gaussian mixture model algorithm, so as to identify multiple user types and create user portraits for multiple user types. The user portraits include typical characteristics, common behavior patterns and potential needs.

[0011] Permission granting unit: Assigns corresponding permission levels and service scopes to different user types based on the login channel. According to the user portrait, the login channel is configured to the corresponding user type to obtain user permission data.

[0012] Dynamic permission update unit: used to continuously observe user activities and historical search data, obtain behavioral trends of changes in permission requirements, use real-time data analysis technology to evaluate users' current needs and interests to obtain change data, and evaluate whether user permission data exceeds the preset standard based on the change data. If so, update the user permission data; otherwise, continue to maintain the user's current permissions.

[0013] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the user data analysis unit specifically includes: Extraction subunit: extract key features from user learning data.

[0014] Expectation step calculation subunit: defines that the data points of the Gaussian mixture model are generated by multiple Gaussian distributions, initializes the parameters, and uses the expectation step algorithm to calculate the posterior probability that the data points belong to each Gaussian component.

[0015] Update subunit: Update the parameters of the mixture components of the Gaussian distribution, including updating the weights, updating the mean, and updating the covariance.

[0016] Iteration subunit: Iteratively execute the desired step calculation subunit and update subunit until the preset maximum number of iterations is reached.

[0017] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the formula for the posterior probability is expressed as:

[0018] In the formula, is the index of the Gaussian component, is the ith data point, is the prior probability of the Gaussian component, is the likelihood probability of the Gaussian component, is the marginal probability.

[0019] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the dynamic authority updating unit specifically includes: Data cleaning and labeling subunit: remove outliers and erroneous data from historical search data, label missing data, and convert different types of data into a unified dimension.

[0020] Feature weight allocation subunit: extracts multiple permission features related to permission requirements from user activities and historical search data, and allocates multiple permission features according to preset weights to obtain multiple feature coefficients.

[0021] Rating calculation subunit: Calculate the user's current needs and user interests based on the feature coefficients to obtain a comprehensive score.

[0022] Rating change analysis subunit: Calculate the comprehensive score according to the preset frequency, and compare the comprehensive score of the previous week with the comprehensive score of the current week to obtain the change data.

[0023] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the model construction and application module includes: Personalized model building unit: extract joint features based on the resource characteristics of the teaching resource model, the user operation scope of the user authority data, and the user habits of the historical search data, select the non-negative matrix decomposition algorithm to calculate the joint features to obtain the training label data, and build a personalized learning model based on the training label data.

[0024] Resource platform optimization unit: used to pre-process historical search data and input it into the personalized learning model, and use collaborative filtering algorithms to sort the resources within the personalized learning model and predict users' future content to optimize the search performance of the personalized learning model.

[0025] Learning plan generation unit: extract key information of user learning data according to input requirements to obtain progress data, standardize the progress data and input it into the personalized learning model to generate customized learning plan resource data.

[0026] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the resource platform optimization unit includes: Data processing subunit: remove errors and invalid search records in historical search data to obtain standard data.

[0027] Algorithm application subunit: Construct a user resource matrix based on standard data, use cosine similarity to calculate the similarity between different users to obtain similarity values, and sort and predict the resources in the personalized learning model based on the similarity values ​​to obtain resource recommendation data.

[0028] Platform update subunit: updates resource recommendation data in real time based on users’ real-time behavior, and updates collaborative filtering algorithms at preset intervals.

[0029] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the personalized model building unit includes: Feature extraction unit: used to extract resource features from the teaching resource model, obtain usage permissions from user permission data analysis, and obtain search habits from historical search data analysis, and combine resource features, usage permissions and search habits into a feature matrix.

[0030] Calculation and construction unit: used to input the feature matrix into the non-negative matrix decomposition algorithm for decomposition, extract potential factors, and thus calculate and generate training label data, and build a personalized learning model based on the training label data.

[0031] According to a digital teaching resource retrieval management system based on online teaching provided by the present invention, the feedback driven optimization module includes: Feedback collection and analysis unit: used to set up feedback optimization channels on the teaching resource model, collect feedback data from users after using the personalized learning model, clean and classify it, analyze user feedback trends and common problems to obtain feedback reports.

[0032] Platform real-time optimization unit: used to determine the optimization direction based on feedback reports, adjust the personalized learning model, and update the resources within the teaching resource model.

[0033] On the other hand, the present invention also provides a method for managing the retrieval of digital teaching resources based on online teaching, and the management method includes: S1: Collect user learning data, teaching resources and historical search data in real time, classify teaching resources according to preset categories to obtain resource statistics, and build a teaching resource model based on resource statistics.

[0034] S2: Extract user types from user learning data according to the Gaussian mixture model algorithm at a preset period, create a login channel according to the user type, grant permissions to each user type according to the login channel to obtain user permission data, and update the user permission data based on user activities and historical search data.

[0035] S3: Build a personalized learning model based on the teaching resource model, user authority data and historical search data, input the historical search data into the personalized learning model for training, optimize the search performance of the personalized learning model, input the user learning data into the personalized learning model, and obtain customized learning planning resource data.

[0036] S4: Set up a feedback optimization channel in the teaching resource model. Users will get feedback data after using the personalized learning model, and optimize the teaching resource model in real time based on the feedback data.

[0037] The present invention provides a digital teaching resource retrieval management system and method based on online teaching. Through dynamic authority management, the user's authority can be adjusted in time according to the user's historical search data and activity dynamics, thereby improving the security and flexibility of the system. Through the establishment of a personalized learning model, the user's learning data can be analyzed in real time, and the learning plan can be customized according to the user's needs, and the retrieval and recommendation of resources can be optimized to improve the utilization and accuracy of teaching resources. At the same time, the real-time optimization mechanism based on user feedback enables the system to maintain synchronous development with user needs and the teaching environment, solves the problem that the system cannot adapt to changes, and ensures that the system can provide users with high-quality teaching resource retrieval management services for a long time.

[0038] The present invention provides a digital teaching resource retrieval management system and method based on online teaching, which builds accurate portraits based on user behavior data, provides customized learning plans and services, meets individual differences, improves learning efficiency and effects, and enhances user satisfaction and loyalty. At the same time, it can also adjust permissions and optimize platforms based on user activities and historical search data, respond to changes in roles, interests, etc. in the learning process in a timely manner, and continuously improve based on feedback, maintain the flexibility and advancement of the system, and adapt to the ever-changing teaching environment and user needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 This is one of the flow charts of a digital teaching resource retrieval management system based on online teaching provided by an embodiment of the present invention; Figure 2 It is a unit flow diagram of the rights management module provided by an embodiment of the present invention; Figure 3 It is a unit flow diagram of the model construction and application module provided in an embodiment of the present invention; Figure 4 The present invention provides a schematic diagram of a method for managing digital teaching resources based on online teaching. DETAILED DESCRIPTION

[0041] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0042] Combine the following Figure 1-Figure 4 The present invention describes a digital teaching resource retrieval management system and method based on online teaching.

[0043] like Figure 1 As shown, an embodiment of the present invention provides a digital teaching resource retrieval management system based on online teaching, including: Resource collection and construction module: used to collect user learning data, teaching resources and historical search data in real time, classify teaching resources according to preset categories to obtain resource statistics, and build a teaching resource model based on resource statistics.

[0044] Permission management module: used to extract user types from user learning data according to the Gaussian mixture model algorithm at a preset period, create login channels according to user types, grant permissions to each user type according to the login channels to obtain user permission data, and update user permission data according to user activities and historical search data.

[0045] like Figure 2 As shown, the rights management module includes: User data analysis unit: used to pre-process user learning data. User learning data A includes (a1, a2, a3, ..., an), where A1 is learning time, A2 is access frequency, A3 is interaction status, and An is test score. The pre-processed user learning data is analyzed by Gaussian mixture model algorithm to identify multiple user types and create user portraits for multiple user types. The user portraits include typical characteristics, common behavior patterns, and potential needs.

[0046] The user data analysis unit specifically includes: Extraction subunit: Extract key features from user learning data. Key features B include (b1, b2, ...bn), b1 is the learning time, b2 is the search frequency, and bn is the search keyword.

[0047] Expected step calculation subunit: defines that the data points of the Gaussian mixture model are generated by multiple Gaussian distributions, initializes the parameters, and uses the expected step algorithm to calculate the posterior probability that the data points belong to each Gaussian component. The formula is expressed as:

[0048] In the formula, is the index of the Gaussian component, is the ith data point, is the prior probability of the Gaussian component, is the likelihood probability of the Gaussian component, is the marginal probability.

[0049] Update subunit: Update the parameters of the mixture components of the Gaussian distribution, including updating the weights, updating the mean, and updating the covariance.

[0050] Update weights: Equal to all sample points Sum from 1 to N, where each term is the probability that sample point i belongs to the kth mixture component , and then divided by the total number of samples N.

[0051] Update the mean: It is equal to the sum of all sample points i from 1 to N, where each item is the probability that the sample point i belongs to the kth mixture component. Multiply by the sample points , divided by the probability that the sample point i belongs to the kth mixture component, which is the sum of all sample points i from 1 to N .

[0052] Update the covariance: It is equal to the sum of all sample points i from 1 to N, where each item is the probability that the sample point i belongs to the kth mixture component. Multiply by the sample points Subtract the mean With sample points Subtract the mean The transpose of , and then divided by the probability of sample point i belonging to the kth mixture component summed over all sample points i from 1 to N .

[0053] Iteration subunit: Iteratively execute the desired step calculation subunit and update subunit until the preset maximum number of iterations is reached.

[0054] Permission granting unit: Based on the login channel, the corresponding permission level and service scope are assigned to different user types. According to the user portrait, the login channel is configured to the corresponding user type to obtain user permission data. For student users, their permission level is mainly set on basic learning functions such as resource browsing, basic course learning, online homework submission and correction result viewing. The service scope covers a rich course resource library of various grades and subjects from elementary school to university, as well as supporting after-class exercises and tutoring materials to meet their daily learning needs; for teacher users, the permission level is upgraded to include teaching management functions such as resource uploading, modification, and deletion, viewing and analyzing student learning progress and grades, and creating and managing class courses. The service scope not only includes a large amount of teaching material resources, but also supports personalized teaching activities, such as online live teaching, creating exclusive teaching classes, and assigning targeted homework; and administrator users are granted the highest level of system permissions, covering comprehensive management functions such as system settings, user information management, data backup and recovery, and overall platform performance monitoring and optimization. The service scope involves the operation, maintenance and security of the entire platform to ensure stable and efficient operation of the system. Configure each type of login channel to the corresponding user type, so as to obtain complete and adapted user authority data in a comprehensive and detailed manner, ensure that each user can obtain operating permissions and service experience that matches their identity and needs on the platform, and promote the orderly and efficient operation of the entire digital teaching resource management system.

[0055] Dynamic permission update unit: used to continuously observe user activities and historical search data, obtain behavioral trends of changes in permission requirements, use real-time data analysis technology to evaluate users' current needs and interests to obtain change data, and evaluate whether user permission data exceeds the preset standard based on the change data. If so, update the user permission data; otherwise, continue to maintain the user's current permissions.

[0056] The dynamic permission update unit includes: Data cleaning and marking subunit: remove outliers and erroneous data from historical search data, mark missing data, and convert different types of data into a unified dimension. For data points that obviously deviate from the normal search frequency range, such as high-frequency search records that appear in a very short period of time, if they do not conform to normal human operating logic, they may be caused by system failures or malicious brushing behaviors, and should be resolutely eliminated; at the same time, for those search data with incorrect formats, unparseable, or seriously mismatched with existing data types on the platform, they should also be carefully identified and cleaned to ensure the accuracy and reliability of the data. When it is found that key information is missing in certain search records, such as missing search time, partial loss of search keywords, or incomplete user identification, these missing locations should be clearly marked with specific symbols or codes in a timely manner, so that they can be clearly identified in the subsequent data analysis stage and appropriate filling strategies can be selected according to the specific situation, or the special impact of these missing values ​​can be directly considered in the relevant model, rather than simply ignoring them, thereby avoiding deviations in analysis results caused by missing data.

[0057] Assign feature weight subunit: Extract multiple permission features related to permission requirements from user activities and historical search data. Permission features include resource download frequency, resource upload frequency, professional depth of search resources, and frequency of editing operations on teaching resources. And assign multiple permission features according to preset weights to obtain multiple feature coefficients. The calculation process of the feature coefficient is: calculate the weighted sum of the permission feature in all data points (weight multiplied by the feature value of each data point, and then summed), and then divide it by the sum of all values ​​of the feature. The feature coefficient obtained in this way comprehensively considers factors such as the importance of the permission feature in all data and its actual frequency of occurrence.

[0058] Rating calculation subunit: Calculate the user's current needs and user interests based on the feature coefficients to obtain a comprehensive score.

[0059] Rating change analysis subunit: Calculate the comprehensive score according to the preset frequency, and compare the comprehensive score of the previous week with the comprehensive score of the current week to obtain the change data.

[0060] Model building and application module: build a personalized learning model based on the teaching resource model, user authority data and historical search data, input the historical search data into the personalized learning model for training, optimize the search performance of the personalized learning model, input the user learning data into the personalized learning model, and obtain customized learning planning resource data.

[0061] like Figure 3 As shown, the model building and application modules include: Personalized model building unit: extract joint features based on the resource characteristics of the teaching resource model, the user operation scope of the user authority data, and the user habits of the historical search data, select the non-negative matrix decomposition algorithm to calculate the joint features to obtain the training label data, and build a personalized learning model based on the training label data.

[0062] The personalized model building unit includes: Feature extraction unit: used to extract resource features from the teaching resource model, obtain usage permissions from user permission data analysis, and obtain search habits from historical search data analysis, and combine resource features, usage permissions and search habits into a feature matrix. The feature matrix formula is expressed as:

[0063] In the formula, is the feature matrix, is the resource feature matrix, is the usage permissions matrix, It is the search habit matrix.

[0064] Calculation and construction unit: used to input the feature matrix into the non-negative matrix decomposition algorithm for decomposition, extract potential factors, and thus calculate and generate training label data. The formula is expressed as:

[0065] In the formula, is the training label data, is a matrix No. Line elements, It is The weights corresponding to the potential factors are is the number of potential factors.

[0066] Based on the latent factors extracted by non-negative matrix decomposition and the given weight vector, the corresponding training label data is generated for each resource calculation, and a personalized learning model is constructed based on the training label data.

[0067] Resource platform optimization unit: used to pre-process historical search data and input it into the personalized learning model, and use collaborative filtering algorithms to sort the resources within the personalized learning model and predict users' future content to optimize the search performance of the personalized learning model.

[0068] The resource platform optimization unit includes: Data processing subunit: remove errors and invalid search records in historical search data to obtain standard data. For erroneous data, it is necessary to use precise data verification algorithms to strictly screen out garbled search records caused by system failures, repeated or truncated search information caused by network anomalies, and abnormal data entries that do not conform to the platform data format specifications to ensure the accuracy and completeness of the data. As for invalid search records, it is necessary to combine intelligent semantic analysis technology with business rule screening to eliminate those common vocabulary searches that are obviously irrelevant to the platform's teaching resources (such as meaningless single letters, numbers, or common non-teaching high-frequency words), search behavior records that are repeated continuously in a very short period of time and have no substantive meaning due to user misoperation, and records that do not conform to normal search logic due to improper use of the search function. At the same time, it is also necessary to combine the platform's resource classification system and user behavior patterns to exclude those search terms that seem normal but have nothing to do with the teaching content provided by the platform, such as searching for the names of other irrelevant websites, and entertainment words that are contrary to the platform's teaching direction. After such a rigorous denoising process, the disorganized, worthless errors and invalid search records are completely removed from the original data, thereby obtaining standard data that meets the requirements of data analysis and can truly reflect the user's search intentions and behavioral characteristics, laying a solid and reliable data foundation for subsequent user portrait construction, personalized learning model training, and teaching resource optimization and recommendation based on these data, ensuring the efficient operation and accurate decision-making of the entire digital teaching resource management system.

[0069] Algorithm application subunit: Construct a user resource matrix based on standard data, use cosine similarity to calculate the similarity between different users to obtain similarity values, and sort and predict the resources in the personalized learning model based on the similarity values ​​to obtain resource recommendation data.

[0070] Platform update subunit: updates resource recommendation data in real time based on users’ real-time behavior, and updates the collaborative filtering algorithm at preset intervals.

[0071] Learning plan generation unit: extract key information of user learning data according to input requirements to obtain progress data, standardize the progress data and input it into the personalized learning model to generate customized learning plan resource data.

[0072] Feedback-driven optimization module: set up feedback optimization channels in the teaching resource model, obtain feedback data after users use the personalized learning model, and optimize the teaching resource model in real time based on the feedback data.

[0073] The feedback-driven optimization module includes: Feedback collection and analysis unit: used to set up feedback optimization channels on the teaching resource model, collect feedback data from users after using the personalized learning model, clean and classify it, analyze user feedback trends and common problems to obtain feedback reports.

[0074] Platform real-time optimization unit: used to determine the optimization direction based on feedback reports, adjust the personalized learning model, and update the resources within the teaching resource model.

[0075] The present invention provides a digital teaching resource retrieval management system based on online teaching. Through dynamic authority management, the user's authority can be adjusted in time according to the user's historical search data and activity dynamics, thereby improving the security and flexibility of the system. Through the establishment of a personalized learning model, the user's learning data can be analyzed in real time, and the learning plan can be customized according to the user's needs, and the retrieval and recommendation of resources can be optimized to improve the utilization and accuracy of teaching resources. At the same time, the real-time optimization mechanism based on user feedback enables the system to maintain synchronous development with user needs and the teaching environment, solves the problem that the system cannot adapt to changes, and ensures that the system can provide users with high-quality teaching resource retrieval management services for a long time.

[0076] like Figure 4 As shown, based on the same inventive concept, the present invention also includes a digital teaching resource retrieval management method based on online teaching, and the management method includes: S1: Collect user learning data, teaching resources and historical search data in real time, classify teaching resources according to preset categories to obtain resource statistics, and build a teaching resource model based on resource statistics.

[0077] S2: Extract user types from user learning data according to the Gaussian mixture model algorithm at a preset period, create login channels according to user types, grant permissions to each user type according to the login channels to obtain user permission data, and update user permission data according to user activities and historical search data.

[0078] S3: Build a personalized learning model based on the teaching resource model, user authority data and historical search data, input the historical search data into the personalized learning model for training, optimize the search performance of the personalized learning model, input the user learning data into the personalized learning model, and obtain customized learning planning resource data.

[0079] S4: Set up a feedback optimization channel in the teaching resource model. Users will get feedback data after using the personalized learning model, and optimize the teaching resource model in real time based on the feedback data.

[0080] The present invention provides a method for managing the retrieval of digital teaching resources based on online teaching, which builds accurate portraits based on user behavior data, provides customized learning plans and services, meets individual differences, improves learning efficiency and effects, and enhances user satisfaction and loyalty. At the same time, it can also adjust permissions and optimize platforms based on user activities and historical search data, respond to changes in roles, interests, etc. in the learning process in a timely manner, and continuously improve based on feedback, maintain the flexibility and advancement of the system, and adapt to the ever-changing teaching environment and user needs.

[0081] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0082] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A digital teaching resource retrieval management system based on online teaching, characterized in that: include: Resource collection and construction module: used to collect user learning data, teaching resources and historical search data in real time, and classify the teaching resources according to preset categories to obtain resource statistical data; and constructing a teaching resource model based on the resource statistical data; Permission management module: used to extract user types from the user learning data according to the Gaussian mixture model algorithm according to a preset period, create a login channel according to the user type, grant permissions to each user type according to the login channel to obtain user permission data, and update the user permission data according to user activities and the historical search data; Model building and application module: building a personalized learning model based on the teaching resource model, user authority data and historical search data, inputting the historical search data into the personalized learning model for training, optimizing the search performance of the personalized learning model, inputting the user learning data into the personalized learning model, and obtaining customized learning planning resource data; Feedback-driven optimization module: used to set feedback optimization channels and optimize the teaching resource model in real time according to the feedback data obtained after the user uses the personalized learning model.

2. According to claim 1, a digital teaching resource retrieval management system based on online teaching is characterized in that: The rights management module includes: A user data analysis unit is used to pre-process the user learning data, analyze the pre-processed user learning data by using a Gaussian mixture model algorithm, thereby identifying a plurality of user types, and creating user profiles for the plurality of user types, wherein the user profiles include typical features, common behavior patterns, and potential needs; Permission granting unit: based on the login channel, grants corresponding permission levels and service scopes to different user types, and according to the user portrait, configures the login channel to the corresponding user type to obtain user permission data; Dynamic permission update unit: used to continuously observe the user activities and the historical search data, obtain the behavioral trend of changes in permission requirements, use real-time data analysis technology to evaluate the user's current needs and user interests to obtain change data, and evaluate whether the user's permission data exceeds the preset standard based on the change data. If so, update the user's permission data, otherwise continue to maintain the user's current permissions.

3. According to claim 2, a digital teaching resource retrieval management system based on online teaching is characterized in that: The user data analysis unit specifically includes: Extraction subunit: extracting key features from the user learning data; Expected step calculation subunit: defines that the data points of the Gaussian mixture model are generated by multiple Gaussian distributions, initializes the parameters, and uses the expected step algorithm to calculate the posterior probability that the data points belong to each Gaussian component; Update subunit: update the parameters of the mixed components of the Gaussian distribution, including updating weights, updating means and updating covariances; Iteration subunit: iteratively executes the desired step calculation subunit and the update subunit until a preset maximum number of iterations is reached.

4. According to claim 3, a digital teaching resource retrieval management system based on online teaching is characterized in that: The formula for the posterior probability is expressed as: In the formula, is the index of the Gaussian component, is the ith data point, is the prior probability of the Gaussian component, is the likelihood probability of the Gaussian component, is the marginal probability.

5. According to claim 2, a digital teaching resource retrieval management system based on online teaching is characterized in that: The dynamic permission updating unit comprises: Data cleaning and marking subunit: removes outliers and erroneous data from the historical search data, marks missing data, and converts different types of data into a unified dimension; A feature weight allocation subunit: extracting a plurality of permission features related to the permission requirement from the user activity and the historical search data, and allocating the plurality of permission features according to preset weights to obtain a plurality of feature coefficients; Rating calculation subunit: calculating the user's current demand and the user's interest according to the characteristic coefficient to obtain a comprehensive score; Rating change analysis subunit: calculates the comprehensive score according to a preset frequency, and compares the comprehensive score of last week with the comprehensive score of the current week to obtain change data.

6. According to the digital teaching resource retrieval management system based on online teaching according to claim 1, it is characterized in that: The model building and application module includes: Personalized model building unit: extracting joint features based on resource features of the teaching resource model, user operation scope of the user authority data, and user habits of the historical search data, selecting a non-negative matrix decomposition algorithm to calculate the joint features to obtain training label data, and building a personalized learning model according to the training label data; Resource platform optimization unit: used for pre-processing the historical search data and inputting it into the personalized learning model, and using collaborative filtering algorithm to sort the resources in the personalized learning model and predict the user's future content to optimize the search performance of the personalized learning model; Learning plan generating unit: extracting key information of the user learning data according to input requirements to obtain progress data, standardizing the progress data and then inputting it into the personalized learning model to generate customized learning plan resource data.

7. A digital teaching resource retrieval management system based on online teaching according to claim 6, characterized in that: The resource platform optimization unit includes: Data processing subunit: removing errors and invalid search records in the historical search data to obtain standard data; Algorithm application subunit: constructing a user resource matrix according to the standard data, using cosine similarity to calculate the similarity between different users to obtain a similarity value, and sorting and predicting the resources in the personalized learning model according to the similarity value to obtain resource recommendation data; Platform update subunit: based on the real-time behavior of users, update the resource recommendation data in real time and update the collaborative filtering algorithm at preset intervals.

8. A digital teaching resource retrieval management system based on online teaching according to claim 6, characterized in that: The personalized model building unit includes: Feature extraction unit: used to extract the resource features from the teaching resource model, obtain the usage rights from the user rights data analysis, and obtain the search habits from the historical search data analysis, and combine the resource features, usage rights and search habits into a feature matrix; Calculation and construction unit: used to input the feature matrix into the non-negative matrix decomposition algorithm for decomposition, extract potential factors, thereby calculate and generate the training label data, and construct the personalized learning model based on the training label data.

9. According to the digital teaching resource retrieval management system based on online teaching according to claim 1, it is characterized in that: The feedback driven optimization module comprises: Feedback collection and analysis unit: used to set the feedback optimization channel on the teaching resource model, collect the feedback data of users after using the personalized learning model, clean and classify it, analyze the feedback trends and common problems of users to obtain feedback reports; Platform real-time optimization unit: used to determine the optimization direction according to the feedback report, adjust the personalized learning model, and update the resources in the teaching resource model.

10. A method for managing the retrieval of digital teaching resources based on online teaching, which adopts a method for managing the retrieval of digital teaching resources based on online teaching as claimed in any one of claims 1 to 9, characterized in that: S1: collecting user learning data, teaching resources and historical search data in real time, classifying the teaching resources according to preset categories to obtain resource statistical data; and constructing a teaching resource model based on the resource statistical data; S2: extracting user types from the user learning data according to the Gaussian mixture model algorithm at a preset period, creating a login channel according to the user type, granting permissions to each user type according to the login channel to obtain user permission data, and updating the user permission data according to user activities and driven historical search data; S3: constructing a personalized learning model based on the teaching resource model, user authority data and historical search data, inputting the historical search data into the personalized learning model for training, optimizing the search performance of the personalized learning model, and inputting the user learning data into the personalized learning model to obtain customized learning planning resource data; S4: A feedback optimization channel is set in the teaching resource model, and feedback data is obtained after the user uses the personalized learning model, and the teaching resource model is optimized in real time based on the feedback data.

Images