Teacher teaching archive system integrating teaching resource sharing

Abstract

The invention belongs to the technical field of teaching, and provides a teacher teaching archive system fusing teaching resource sharing, which comprises a login module, a resource uploading module, a management auditing module and a retrieval recommendation module, deploys edge computing nodes, compares a pre-stored identity information set with basic information provided by a user, and sends the information to the user; a multi-dynamic verification method is adopted according to network environment conditions to ensure system safety, edge nodes check uploaded resources, the system carries out uploading processing according to the resources with high priorities when uploading congestion occurs, an administrator checks the uploaded teaching resources according to an auditing system and stores the teaching resources by using a distributed storage technology, and the distributed storage technology is used for storing the teaching resources. And combining the importance degree value of the keyword and the similarity value of the keyword, and calculating by utilizing a non-factor linear formula to obtain a final recommendation value, and performing recommendation according to the recommendation value.

Description

Technical Field

[0001] The present invention belongs to the technical field of teaching, and specifically relates to a teacher teaching file system integrating teaching resource sharing. Background Art

[0002] With the continuous development of educational informatization, the digital management of teaching resources has become increasingly important; there are some deficiencies in the traditional teacher teaching file system, such as the single user login verification method, which is prone to illegal user access and difficult to guarantee the system security; during the teaching resource uploading process, there is a lack of unified processing for resources in different formats, which not only wastes storage space but also may cause format incompatibility problems in subsequent use; in addition, in terms of resource retrieval and recommendation, there is a lack of a scientific and reasonable mechanism, and it is difficult for teachers to quickly obtain high-quality teaching resources that meet their own needs.

[0003] Therefore, the present invention provides a teacher teaching file system integrating teaching resource sharing. Summary of the Invention

[0004] In order to make up for the deficiencies of the prior art and solve at least one of the technical problems proposed in the background art.

[0005] The technical solution adopted by the present invention to solve its technical problems is as follows:

[0006] In the first aspect, the present invention provides a teacher teaching file system integrating teaching resource sharing, including:

[0007] Login module: Compare the pre-stored identity information set with the user's basic information, and adopt a multi-dynamic verification method according to the network environment conditions to ensure system security; after passing the verification, allow the user to log in.

[0008] Resource upload module: Based on the upload after logging in, first process the teaching resources and then upload them.

[0009] Management and review module: Based on the processed teaching resources, the administrator reviews the uploaded teaching resources according to the review system, and the teaching resources passing the review adopt the distributed storage technology.

[0010] Retrieval and recommendation module: Based on the retrieval after logging in, reduce the dimension of the keywords, judge the importance degree value of the keywords, combine the keyword similarity value, process to obtain the final recommendation value, and perform retrieval and recommendation according to the final recommendation value.

[0011] In the second aspect, the present invention provides a teacher teaching file method integrating teaching resource sharing, including:

[0012] S1: Compare the pre-stored identity information set with the user's basic information, and adopt a multi-dynamic verification method according to the network environment conditions to ensure system security; after passing the verification, allow the user to log in;

[0013] S2: Based on uploading after logging in, first process the teaching resources and then upload them;

[0014] S3: Based on the processed teaching resources, the administrator audits the uploaded teaching resources according to the audit system, and the teaching resources that pass the audit adopt distributed storage technology;

[0015] S4: Based on retrieval after logging in, reduce the dimension of keywords, judge the importance value of keywords, combine the keyword similarity value, process to obtain the final recommendation value, and perform retrieval recommendation according to the final recommendation value.

[0016] The beneficial effects of the present invention are as follows: Adopt a multi-dynamic verification method, combine basic identity information and biological information for verification, compared with the traditional single-dynamic verification method, effectively reduce illegal user access, and ensure the security and reliability of the system; when users upload educational resources, unify the processing of different types of educational resources, avoiding subsequent problems of inconsistent formats; at the same time, use edge node technology to optimize the situation of upload congestion and slow speed in real time, improve the upload efficiency, the retrieval recommendation module calculates the recommendation value according to the relevant data of educational resources, and performs sorting and recommendation according to the recommendation value, which is convenient for teachers to quickly obtain high-quality teaching resources that meet their own needs and improve the utilization rate of teaching resources. Brief Description of the Drawings

[0017] The present invention will be further described below with reference to the accompanying drawings.

[0018] Figure 1 is a program block diagram of a teacher teaching file system integrating teaching resource sharing according to an embodiment of the present invention;

[0019] Figure 2 is a step flowchart of a teacher teaching file method integrating teaching resource sharing according to an embodiment of the present invention. Detailed Embodiments

[0020] In order to make the technical means, creative features, achieved purposes and effects of the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments.

[0021] Embodiment 1

[0022] As Figure 1 shown, a teacher teaching file system integrating teaching resource sharing according to an embodiment of the present invention includes:

[0023] Login Module: First, collect and pre-store the basic information of teachers and administrators, including name, account and password, age, and affiliated teaching institution; denote it as the set of basic identity information A = {S1, S2, S3, S4}: where S1 is the name of the teacher or administrator, S2 is the account and password, S3 is the age, and S4 is the affiliated educational institution; establish a set of all the basic identity information of teachers and administrators, denoted as the pre-stored identity information set Ai = {i = 1, 2, 3…n}; where n is the number of elements in the pre-stored identity information set; at the same time, collect the biometric information of teachers and administrators, and the biometric information includes: fingerprints, faces;

[0024] The basic identity information has fast verifiability and can quickly log in to the system when the user accesses the system, while the biometric information has a higher level of uniqueness and accuracy;

[0025] The login module adopts a multi-dynamic verification method, which effectively reduces the access of illegal users compared with the single-dynamic verification method in traditional teaching systems, ensuring the security of the system; the specific content is as follows:

[0026] When the user logs in to the system, compare the basic information provided by the user with the pre-stored identity information set for verification; the system traverses each element in the pre-stored identity information set Ai = {i = 1, 2, 3…n}, and conducts the following verification on the elements in the pre-stored identity information set:

[0027] First, set an initial value to express whether the user information is verified; the initial value is set to false: let isValid = false; the system traverses each element in the pre-stored identity information set Ai = {i = 1, 2, 3…n}, and conducts the following verification on the elements in the pre-stored identity information set:

[0028] Verify the name: If S_name = S1;

[0029] Verify the account and password: And S_account_and_password = S2;

[0030] Verify the age: And S_age = S3;

[0031] Verify the affiliated teaching institution: And S_affiliated_teaching_institution = S4;

[0032] Then set isValid to true; otherwise, the value of isValid remains unchanged; only when the above four conditions are met simultaneously will it change; as long as one condition is not met, isValid is false;

[0033] When all the basic information conforms, the system allows login;

[0034] When the basic information contains one or more non-conformances, the system does not allow login;

[0035] Define a risk value for the system. The risk value is a comprehensive indicator for evaluating different risk types;

[0036] The risk types include: data leakage, malicious virus software attack, network attack; divide a weight level for each risk type; calculate the risk value according to the formula F = β1·L1 + β2·L2 + β3·L3; where, β1, β2, β3 are the risk type weight levels; L1 is the data leakage score, L2 is the malicious virus software attack score, and L3 is the network attack score; for each login access of the user, compare and analyze the calculated risk value with the preset risk threshold. It should be noted that the risk threshold is a reference value for the network environment security set by technicians in this industry, and its function is to measure whether there are potential security hazards in the network environment system;

[0037] If the risk value is greater than or equal to the risk threshold; then additionally verify the user's biometric information; if the risk value is less than the risk threshold, it is considered that the current system network environment is secure and the user can log in quickly through the basic identity information;

[0038] By adopting the multi-dynamic verification method, it can effectively reduce the login of illegal users and potential network security hazards, and at the same time improve the user experience;

[0039] Teaching resource upload and processing flow module: When a teacher uploads teaching resources, the edge node first performs local cache check and data preprocessing, and then uploads the processed file to the cloud;

[0040] Specifically, the teacher selects the local teaching resources to be uploaded through the resource upload entry of the system. The resources include: courseware, lesson plans, videos; and fills in and marks the information related to the teaching resources. The related information includes: subject, grade, resource name, introduction; the edge node performs a local cache check. After the edge node receives the upload request, it first checks the local cache;

[0041] If it is found that a file with the same format already exists in the cache, skip the subsequent normalization processing steps and directly enter the link to upload to the cloud;

[0042] If there is no file of this format in the local cache, perform normalization processing on the uploaded file;

[0043] Perform unified operation on video format: Use the FFmpeg video transcoding tool; by configuring transcoding parameters, uniformly convert video files in different formats into MP4 format;

[0044] Unify the operation of audio formats: Use Audacity to unify the audio formats. For audio files in different formats, first read the file content, and then use the built-in format conversion function of Audacity to convert them into the formats specified by the platform.

[0045] Unify the operation of graphic and text formats: For pictures, use the Pillow library of Python. When uploading pictures in different formats, open the pictures through the Pillow library and then save them in the specified format.

[0046] For document graphic and text resources, use the conversion interface of WPS office software to convert them into PDF format to ensure the integrity of the graphic and text content and the consistency of the display effect.

[0047] For large files: Compress large files to reduce the file size, speed up the upload speed, and save network bandwidth and cloud storage costs.

[0048] Upload the processed educational resources.

[0049] When there is an upload congestion, the system sorts according to the priority of the upload tasks. The priority can be determined according to the scarcity of teaching resources and the length of the teacher's educational experience. For high-priority upload tasks, network bandwidth and server resources are preferentially allocated. At the same time, use the system push module to feedback the congestion situation to the teacher and provide the estimated waiting time so that the teacher can understand the upload progress.

[0050] The technical solution of this embodiment is as follows: Adopt the multi-dynamic verification method, combine the basic identity information and biometric information for verification. Compared with the traditional single-dynamic verification method, it effectively reduces the access of illegal users and ensures the security and reliability of the system. When users upload educational resources, the unified processing of different types of educational resources saves a large amount of storage space and avoids the problem of subsequent format inconsistency. Use edge node technology to optimize the upload congestion situation in real time. The system sorts according to the priority of the upload tasks, preferentially allocates network bandwidth and server resources for high-priority tasks, and feedbacks the congestion situation and the estimated waiting time to the teacher through the system push module so that the teacher can understand the upload progress.

[0051] Embodiment 2

[0052] Based on Embodiment 1, the present invention also provides a teacher teaching file system integrating teaching resource sharing, including:

[0053] Management and review module: Based on the teaching resources uploaded by the teacher after logging in, after the management review unit receives the upload information successfully, the management personnel will receive a system prompt message, and the prompt message includes: email, system pop-up window, message push; ensure that the management personnel can know in the first time that there are new teaching resources to be reviewed.

[0054] After the management staff receives the notice, the system will present the basic information of the educational resources to be reviewed for the management staff. The basic information of the educational resources includes: teacher's name, resource name, subject, grade, and upload time, which facilitates the management staff to quickly understand the resource overview;

[0055] The management staff strictly reviews the educational resources uploaded by teachers according to the review system;

[0056] If the uploaded educational resources successfully pass the review system, the administrator adopts the distributed storage technology, classifies the educational resources by grade and subject, stores the classified educational resources, and users can quickly locate the required resources, significantly improving the retrieval efficiency; Distributed storage can dynamically allocate resources, avoid data redundancy, thus maximizing the use of storage space. Classified storage simplifies the data management process, and the administrator can update, back up, and monitor resources more efficiently to ensure that the resources can still be accessed normally when some nodes fail;

[0057] If the review fails, the management staff will explain and illustrate in detail the basis for the failure according to the review system; At the same time, the system will generate a corresponding message indicating the failure of the review and use the system push module to push the message of the failure of the review to the uploading teacher in real time for timely reminder purposes; Improve the uploading efficiency of teaching resources;

[0058] Through the above rigorous review and information improvement process, the quality of the educational resource library is guaranteed and the educational resources are further standardized;

[0059] Retrieval and recommendation module: The teaching resources after passing the review will be stored in the educational resource sharing system for teachers in need to refer to and use; Through latent semantic analysis, the keyword dimension is reduced, and keyword information can be better captured; When a teacher searches for a certain keyword, first calculate the semantic similarity, represent all the key retrieval words as vectors, and then use the similarity calculation method to calculate the importance of the keywords; Use the cosine similarity to calculate the keyword similarity between two keyword vectors, and finally introduce a formula with non-linear factors to calculate the final recommendation value, as follows:

[0060] Decompose the keyword-document matrix into three matrices through singular value decomposition SVD, that is: A = U×∑V T ; where A is the keyword-document matrix, U is the keyword-topic matrix, ∑ is the singular value matrix, and V is the document-topic matrix. In this way, the latent semantic structure in the text is discovered, the keyword dimension is reduced, and the semantic information of the text is better captured;

[0061] Represent the keyword text as a vector. Assume that there are n different keywords in the text and the size of the keyword vocabulary is n. For a document, its vector representation is: Among them, v i represents the frequency of occurrence of the i-th keyword in the document; the frequency of occurrence of the keyword is calculated using the TF-IDF formula, specifically: Among them, n i，j is the number of times the keyword i appears in the document j; the denominator is the sum of the number of occurrences of all words in the document j;

[0062] Next, measure the importance of the keyword in the document and calculate the inverse document frequency. The specific formula is: Among them, N is the total number of documents in the document set, and |D i | is the number of documents containing the keyword i; finally, use the formula: TF-IDF i，j = TF i，j × IDF i ; calculate the keyword importance value TF-IDF i,j ;

[0063] Similarly, use the cosine similarity to calculate the similarity value between two vectors. The specific formula is: Among them, and represent the vectors of two keywords, and the value range is between [-1, 1]. The closer the value is to 1, the more similar the two vectors are, and the keyword similarity value is calculated;

[0064] Finally, use the non-factor linear formula to calculate the final recommendation value;

[0065] Specifically, use the non-factor linear formula: Calculate the final recommendation value, where I is the keyword similarity value; S is the keyword importance value; w1 and w2 are preset proportionality coefficients;

[0066] Sort the calculated recommendation values from largest to smallest. The largest recommendation value is set at the position of the first row on the first page of the system recommendation module, and the recommendation values are recommended in order from largest to smallest according to the above method;

[0067] The technical solution of this embodiment is as follows: After the teacher uploads teaching resources, the administrator strictly reviews them according to the review system. If the resources pass the review, distributed storage technology is used to store them classified by grade or subject, which can dynamically allocate resources, avoid data redundancy, improve retrieval efficiency, and ensure that the resources can still be accessed when some nodes fail; the singular value decomposition (SVD) is used to reduce the keyword dimension of the keyword-document matrix, represent the keyword text as a vector, calculate the occurrence frequency of the keyword in the document using the TF-IDF formula, calculate the similarity of two keyword vectors using the cosine similarity, and calculate the final recommendation value using the non-linear factor formula; sort the recommendation values from largest to smallest, and place the one with the largest recommendation value in the first row of the first page of the system recommendation module, and recommend them in turn.

[0068] Embodiment 3

[0069] As Figure 2 shown, based on Embodiment 1 and Embodiment 2, the present invention also provides a method for a teacher's teaching file integrating teaching resource sharing, including:

[0070] S1: First, collect and pre-store the basic information of teachers and administrators, including name, account and password, age, and affiliated teaching institution; denote it as the basic identity information set A = {S1, S2, S3, S4}: where S1 is the name of the teacher or administrator, S2 is the account and password, S3 is the age, and S4 is the affiliated educational institution; establish a set of all the basic identity information sets of teachers and administrators, denoted as the pre-stored identity information set Ai = {i = 1, 2, 3...n}; where n is the number of elements in the pre-stored identity information set.

[0071] At the same time, collect the biometric information of teachers and administrators, and the biometric information includes: fingerprints, faces; the system traverses each element in the pre-stored identity information set Ai = {i = 1, 2, 3...n} and verifies the elements in the pre-stored identity information set: if the verification passes, the system allows login.

[0072] Define a risk value for the system, and the risk value is a comprehensive index for evaluating different risk types.

[0073] For each login access of the user, compare and analyze the calculated risk value with the preset risk threshold. It should be noted that the risk threshold is a reference value for the network environment security set by those skilled in the art, and its function is to measure whether there are potential security hazards in the network environment system.

[0074] If the risk value is greater than or equal to the risk threshold; then additionally verify the user's biometric information; if the risk value is less than the risk threshold, it is considered that the current system network environment is secure and can be quickly logged in through the basic identity information.

[0075] S2: When the teacher uploads teaching resources, the edge node first conducts local cache checks and data preprocessing, and then uploads the processed file to the cloud. For the local cache check of the edge node, after receiving the upload request, the edge node first checks the local cache. If it is found that a file of the same format already exists in the cache, the subsequent unified processing steps are skipped, and it directly enters the link of uploading to the cloud. If there is no file of this format in the local cache, the uploaded file is subjected to unified processing.

[0076] Unified operation for video format: Use the FFmpeg video transcoding tool; by configuring transcoding parameters, uniformly convert video files of different formats into MP4 format.

[0077] Unified operation for audio format: Use Audacity to unify the audio format; for audio files of different formats, first read the file content, and then use the built-in format conversion function of Audacity to convert it into the format specified by the platform.

[0078] Unified operation for graphic and text format: For pictures, use the Pillow library in Python; when uploading pictures of different formats, open the pictures through the Pillow library and then save them in the specified format.

[0079] For document graphic and text resources, use the conversion interface of WPS office software to convert them into PDF format to ensure the integrity of the graphic and text content and the consistency of the display effect.

[0080] For large files: Compress large files to reduce the file size, speed up the upload speed, save network bandwidth and cloud storage costs.

[0081] S3: Based on the teaching resources uploaded by the teacher after logging in, the administrator strictly reviews the educational resources uploaded by the teacher according to the review system. If the uploaded educational resources successfully pass the review system, the administrator adopts the distributed storage technology, classifies the educational resources by grade and subject, stores the classified educational resources, and users can quickly locate the required resources, significantly improving the retrieval efficiency. Distributed storage can dynamically allocate resources, avoid data redundancy, thereby maximizing the use of storage space. Classified storage simplifies the data management process, and the administrator can update, back up and monitor resources more efficiently to ensure that the resources can still be accessed normally when some nodes fail.

[0082] If the review fails, the administrator will, according to the review system, explain and illustrate in detail the basis for the failure of the review. At the same time, the system will generate a corresponding message of failing the review and use the system push module to push the message of failing the review to the uploading teacher in real time for timely reminder purposes, so as to improve the upload efficiency of teaching resources.

[0083] S4: By performing latent semantic analysis to reduce the keyword dimension, it is possible to better capture keyword information. When a teacher searches for a certain keyword, first calculate the semantic similarity, represent all the key retrieval words as vectors, and then use the similarity calculation method to calculate the importance value of the keyword. Use the cosine similarity to calculate the keyword similarity value between two keyword vectors, and finally introduce a formula with non-linear factors to calculate the final recommendation value.

[0084] Decompose the keyword-document matrix into three matrices through singular value decomposition (SVD), namely: A = U × ∑V T ; where A is the keyword-document matrix, U is the keyword-topic matrix, ∑ is the singular value matrix, and V is the document-topic matrix. In this way, discover the latent semantic structure in the text, reduce the keyword dimension, and better capture the semantic information of the text.

[0085] Represent the keyword text as a vector. Assume there are n different keywords in the text and the size of the keyword vocabulary is n. For a document, its vector representation is: where, v i represents the frequency of occurrence of the i-th keyword in the document; use the TF-IDF formula to calculate the frequency of occurrence of the keyword, specifically: where, n i，j is the number of times keyword i appears in document j; the denominator is the sum of the number of occurrences of all words in document j;

[0086] Next, measure the importance of the keyword in the document and calculate the inverse document frequency. The specific formula is: where, N is the total number of documents in the document set, |D i | is the number of documents containing keyword i; finally use the formula: TF-IDF i，j = TF i，j × IDF i ; calculate the importance value of the keyword TF-IDF i,j ;

[0087] Similarly, use the cosine similarity to calculate the similarity value between two vectors. The specific formula is: where, and represent the vectors of two keywords, and the value range is between [-1, 1]. The closer the value is to 1, the more similar the two vectors are. Calculate the keyword similarity value;

[0088] Finally, use a non-factor linear formula to calculate the final recommendation value;

[0089] Specifically, use the non-factor linear formula: The final recommended value is calculated, where I is the keyword similarity value; S is the keyword importance value; w1 and w2 are preset proportionality coefficients;

[0090] The calculated recommended values are sorted from largest to smallest, where the largest recommended value is set at the position of the first row on the first page of the system recommendation module, and the recommended values are recommended in order from largest to smallest according to the above method;

[0091] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A teacher's teaching file system integrating teaching resource sharing, characterized in that: It includes: Login module: Compare the pre-stored identity information set with the user's basic information, and adopt a multi-dynamic verification method according to the network environment conditions to ensure system security; Allow the user to log in after verification; Resource upload module: Based on the upload after login, first process the teaching resources and then upload them; Management and review module: Based on the processed teaching resources, the administrator reviews the uploaded teaching resources according to the review system, and the reviewed teaching resources adopt distributed storage technology; Retrieval and recommendation module: Based on the retrieval after login, reduce the keyword dimension, judge the importance value of the keyword, combine the keyword similarity value, process to obtain the final recommendation value, and perform retrieval and recommendation according to the final recommendation value.

2. The teacher teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process of comparing the pre-stored identity information set with the user's basic information is as follows: Compare and verify the user's basic information with the pre-stored identity information set; The system traverses each element in the pre-stored identity information set for verification. If all basic information conforms, the system allows login.

3. The teacher teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process of adopting a multi-dynamic verification method according to the network environment conditions to ensure system security is as follows: For each login access of the user, compare and analyze the calculated risk value with the preset risk threshold. The risk threshold is a reference value for network environment security set by technicians in this industry, and its function is to measure whether there are potential security hazards in the network environment system; If the risk value is greater than or equal to the risk threshold, then additionally verify the user's biological information.

4. The teacher's teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process of first processing the teaching resources and then uploading them is as follows: For videos: Use the FFmpeg video transcoding tool; Uniformly convert video files in different formats to the MP4 format; For audio operations: Use Audacity to unify the audio format; First read the file content, and convert it to the format specified by the platform through the built-in format conversion function of Audacity; For graphic operations: Use the Pillow library of Python; Open the picture through the Pillow library and then save it in the specified format; For document graphic resources, use the conversion interface of WPS office software to convert them to the PDF format.

5. The teacher's teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process of adopting distributed storage technology for the reviewed teaching resources is as follows: Classify and process the educational resources by grade and subject, and store the classified and processed educational resources.

6. The teacher's teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process of first reducing the keyword dimension is as follows: Decompose the keyword-document matrix into three matrices, namely, keyword-document matrix, keyword-topic matrix, and document-topic matrix; Reduce the keyword dimension.

7. The teacher teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process for determining the importance value of the judgment keyword is as follows: First, use the TF-IDF formula to calculate the frequency of the keyword appearance. Then, measure the importance of the keyword in the document and calculate the inverse document frequency. Finally, use the formula: TF-IDF i，j = TF i，j × IDF i ; Calculate the keyword importance.

8. A teacher's teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific calculation process of the keyword similarity value is as follows: Use cosine similarity to calculate the similarity between two vectors to obtain the keyword similarity.

9. The teacher's teaching file system integrating teaching resource sharing according to claim 1, characterized in that: The specific process of obtaining the final recommended value through the processing is as follows: Using the non-factor linear formula: The final recommended value is calculated, where I is the keyword similarity; S is the keyword importance; and w1 and w2 are preset proportionality coefficients.

10. A method for teachers' teaching archives integrating teaching resource sharing, characterized in that: It includes: S1: Compare the pre-stored identity information set with the user's basic information, and adopt a multi-dynamic verification method according to the network environment conditions to ensure system security; Allow the user to log in after verification; S2: Based on the upload after login, first process the teaching resources and then upload them; S3: Based on the processed teaching resources, the administrator reviews the uploaded teaching resources according to the review system, and the reviewed teaching resources adopt distributed storage technology; S4: Based on the retrieval after login, reduce the dimension of keywords, judge the importance value of keywords, combine the keyword similarity value, process to obtain the final recommendation value, and perform retrieval recommendation according to the final recommendation value.

Images