Courseware auditing and duplicate checking method and system based on multi-modal fusion

Abstract

The invention discloses a courseware auditing and duplicate checking method and system based on multi-modal fusion, and is applied to the technical field of courseware auditing. The method comprises the following steps: classifying and pre-checking courseware to be uploaded, extracting keywords, and comparing the keywords with pre-classified labels of the courseware; if the keywords are inconsistent with the pre-classified labels of the courseware, course labels are modified, and if the keywords are consistent with the pre-classified labels of the courseware, duplicate checking analysis is carried out on the courseware to be uploaded; dividing the courseware to be uploaded into a multi-modal data form, and comparing the multi-modal data form with the corresponding content in the courseware resource pool; the repetition rate of the courseware to be uploaded is calculated, and if the repetition rate is lower than a set threshold value, it is judged that the audit of the courseware to be uploaded is passed. According to the method, whether course pre-classification is correct or not can be detected, and multi-mode courseware pre-classification processing is achieved, so that the situation that course content comparison granularities are inconsistent is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of courseware review, and more specifically to a courseware review and duplicate checking method and system based on multimodal fusion. Background Art

[0002] With the development of Internet technology, online courseware has taken on multiple forms such as video, audio, text, and pictures, which has increased the difficulty of courseware review and duplication checking. When teachers upload courseware to the State Grid Academy system, the following problems often occur: first, it is impossible to verify whether the labels or classifications defined by the author for the courseware are accurate. Errors in courseware classification will affect the construction of the knowledge system and burden the maintenance of course resources; second, there is a lack of a complete courseware duplication checking mechanism, which has the risk of duplicate construction of course resources, resulting in knowledge redundancy and occupying platform resources; third, multimodal data such as video, audio, text, and images have not formed a unified resource pool, which cannot effectively support high-precision course resource analysis; fourth, the course review and duplication checking process relies heavily on manual verification, has a low fault tolerance rate, and cannot be automated. Therefore, how to provide a courseware review and duplication checking method and system based on multimodal fusion is an urgent problem that technicians in this field need to solve. Summary of the invention

[0003] In view of this, the present invention provides a courseware review and duplication checking method and system based on multimodal fusion, adopts a multimodal separation detection method, and combines artificial intelligence algorithms to create a new mode of courseware review and duplication checking.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A courseware review and duplicate checking method based on multimodal fusion, comprising the following steps:

[0006] S1. Classify and pre-examine the courseware to be uploaded, extract keywords based on the content of the courseware to be uploaded, and compare the keywords with the labels of the pre-classification of the courseware;

[0007] S2. If the keywords are inconsistent with the labels of the pre-classified courseware, modify the course labels until the keywords are consistent with the labels of the pre-classified courseware. If the keywords are consistent with the labels of the pre-classified courseware, perform a duplicate check on the courseware to be uploaded.

[0008] S3, splitting the courseware to be uploaded into multimodal data forms and comparing them with the corresponding contents in the courseware resource pool;

[0009] S4. Calculate the repetition rate of the courseware to be uploaded. If the repetition rate is higher than the set threshold, the courseware to be uploaded is judged to have a high repetition rate and will not be put online. If the repetition rate is lower than the set threshold, the courseware to be uploaded is judged to have passed the review.

[0010] Optionally, in S1, the keyword proportion is calculated based on the statistical word frequency, and the phrases with a keyword proportion greater than a threshold are taken as keywords.

[0011] Optional, S3 is:

[0012] S31, classifying the data types of the contents in the uploaded courseware;

[0013] S32, converting the non-text data content of the uploaded courseware into text data content;

[0014] S33. Generate an overview of the text data content using natural language processing.

[0015] Optionally, S4 is specifically:

[0016] S41, comparing the text data content summary of the uploaded courseware with the corresponding content in the courseware resource pool and calculating the summary repetition rate;

[0017] S42, if the summary repetition rate is lower than the set threshold, the courseware to be uploaded is judged as non-repetitive, and if the summary repetition rate is higher than the set threshold, the original text repetition rate is continued to be calculated;

[0018] S43, if the repetition rate of the original text is lower than the set threshold, the courseware to be uploaded is judged as non-repetitive; if the repetition rate of the original text is higher than the set threshold, the courseware is judged as repetitive;

[0019] S44, traversing the courseware resource pool, if the text data content of the courseware to be uploaded is not repeated in all the stock resources, it is determined that the courseware to be uploaded has passed the review.

[0020] Optionally, the data types of the content in the courseware include: video, audio, text, PPT, and pictures.

[0021] A courseware review and duplication checking system based on multimodal fusion executes the above-mentioned courseware review and duplication checking method based on multimodal fusion, including a classification pre-review module, a keyword comparison module, a content comparison module and an online determination module, wherein the classification pre-review module is used to perform classification pre-review on the courseware to be uploaded, the keyword comparison module is used to compare the keywords with the pre-classified labels of the courseware, the content comparison module is used to split the courseware to be uploaded into multimodal data forms and compare them with the corresponding content in the courseware resource pool, and the online determination module is used to determine whether the courseware to be uploaded has passed the review.

[0022] It can be seen from the above technical solution that compared with the prior art, the present invention provides a method and system for courseware review and duplication checking based on multimodal fusion, which has the following beneficial effects: the present invention dynamically manages course labels and can detect whether the course pre-classification is correct; the present invention performs review and duplication checking before the course is uploaded, which can reduce the workload of the State Grid Academy resource development administrator; the present invention can realize multimodal courseware pre-classification processing to avoid inconsistent granularity of course content comparison; the present invention has an automated processing mechanism, courseware labels can realize real-time self-detection and correction, courseware review and duplication checking can be completed independently, and can also assist courseware reviewers in their work. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in 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 described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0024] Figure 1 It is a flow chart of the courseware review and duplicate checking method of the present invention;

[0025] Figure 2 The present invention is a flow chart of courseware content comparison and duplicate checking rate calculation. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] The embodiment of the present invention discloses a courseware review and duplication checking method based on multimodal fusion, such as Figure 1 As shown, the following steps are included:

[0028] S1. Classify and pre-examine the courseware to be uploaded, extract keywords based on the content of the courseware to be uploaded, and compare the keywords with the labels of the pre-classification of the courseware;

[0029] S2. If the keywords are inconsistent with the labels of the pre-classified courseware, modify the course labels until the keywords are consistent with the labels of the pre-classified courseware. If the keywords are consistent with the labels of the pre-classified courseware, perform a duplicate check on the courseware to be uploaded.

[0030] S3, splitting the courseware to be uploaded into multimodal data forms and comparing them with the corresponding contents in the courseware resource pool;

[0031] S4. Calculate the repetition rate of the courseware to be uploaded. If the repetition rate is higher than the set threshold, the courseware to be uploaded is judged to have a high repetition rate and will not be put online. If the repetition rate is lower than the set threshold, the courseware to be uploaded is judged to have passed the review.

[0032] Furthermore, in S1, the keyword proportion is calculated based on the statistical word frequency, and the phrases with a keyword proportion greater than a threshold are taken as keywords.

[0033] In the embodiment of the present invention, the labels of the pre-classified courseware are uploaded by the courseware maker.

[0034] Further, such as Figure 2 As shown, S3 is specifically:

[0035] S31, classifying the data types of the contents in the uploaded courseware;

[0036] S32, converting the non-text data content of the uploaded courseware into text data content;

[0037] S33. Generate an overview of the text data content using natural language processing.

[0038] In the embodiment of the present invention, S32 uses OCR technology to convert the image into text, and S33 uses TF-IDF-based text summarization technology to generate an overview of the text data content.

[0039] Further, such as Figure 2 As shown, S4 is specifically:

[0040] S41, comparing the text data content summary of the uploaded courseware with the corresponding content in the courseware resource pool and calculating the summary repetition rate;

[0041] S42, if the summary repetition rate is lower than the set threshold, the courseware to be uploaded is judged as non-repetitive, and if the summary repetition rate is higher than the set threshold, the original text repetition rate is continued to be calculated;

[0042] S43, if the repetition rate of the original text is lower than the set threshold, the courseware to be uploaded is judged as non-repetitive; if the repetition rate of the original text is higher than the set threshold, the courseware is judged as repetitive;

[0043] S44, traversing the courseware resource pool, if the text data content of the courseware to be uploaded is not repeated in all the stock resources, it is determined that the courseware to be uploaded has passed the review.

[0044] In the embodiment of the present invention, the cosine similarity based on TF-IDF word vectors is used to calculate the repetition rate between texts.

[0045] Furthermore, the data types of the content in the courseware include: video, audio, text, PPT, and pictures.

[0046] and Figure 1 Corresponding to the method described, an embodiment of the present invention also discloses a courseware review and duplication checking system based on multimodal fusion, which executes the above-mentioned courseware review and duplication checking method based on multimodal fusion, including a classification pre-review module, a keyword comparison module, a content comparison module and an online judgment module, wherein the classification pre-review module is used to perform classification pre-review on the courseware to be uploaded, the keyword comparison module is used to compare the keywords with the pre-classified labels of the courseware, the content comparison module is used to split the courseware to be uploaded into multimodal data forms and compare them with the corresponding content in the courseware resource pool, and the online judgment module is used to determine whether the courseware to be uploaded has passed the review.

[0047] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0048] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A courseware review and duplication checking method based on multimodal fusion, characterized in that: The following steps are involved: S1. Classify and pre-examine the courseware to be uploaded, and compare the keywords with the pre-classified labels of the courseware according to the content of the courseware to be uploaded; S2. If the keywords are inconsistent with the labels of the pre-classified courseware, modify the course labels until the keywords are consistent with the labels of the pre-classified courseware. If the keywords are consistent with the labels of the pre-classified courseware, perform a duplicate check on the courseware to be uploaded. S3, splitting the courseware to be uploaded into multimodal data forms and comparing them with the corresponding contents in the courseware resource pool; S4. Calculate the repetition rate of the courseware to be uploaded. If the repetition rate is higher than the set threshold, the courseware to be uploaded is judged to have a high repetition rate and will not be put online. If the repetition rate is lower than the set threshold, the courseware to be uploaded is judged to have passed the review.

2. According to a method for courseware review and duplication checking based on multimodal fusion according to claim 1, it is characterized in that: In S1, the keyword proportion is calculated based on the statistical word frequency method, and the phrases with a keyword proportion greater than a threshold are regarded as keywords.

3. According to a method for courseware review and duplication checking based on multimodal fusion according to claim 1, it is characterized in that: S3 is specifically: S31, classifying the data types of the contents in the uploaded courseware; S32, converting the non-text data content of the uploaded courseware into text data content; S33. Generate an overview of the text data content using natural language processing.

4. According to the method for courseware review and duplication checking based on multimodal fusion according to claim 3, it is characterized in that: S4 is specifically: S41, comparing the text data content summary of the uploaded courseware with the corresponding content in the courseware resource pool and calculating the summary repetition rate; S42, if the summary repetition rate is lower than the set threshold, the courseware to be uploaded is judged as non-repetitive, and if the summary repetition rate is higher than the set threshold, the original text repetition rate is continued to be calculated; S43, if the repetition rate of the original text is lower than the set threshold, the courseware to be uploaded is judged as non-repetitive; if the repetition rate of the original text is higher than the set threshold, the courseware is judged as repetitive; S44, traversing the courseware resource pool, if the text data content of the courseware to be uploaded is not repeated in all the stock resources, it is determined that the courseware to be uploaded has passed the review.

5. According to the method of courseware review and duplication checking based on multimodal fusion according to claim 3, it is characterized in that: The data types of content in the courseware include: video, audio, text, PPT, and pictures.

6. A courseware review and duplicate checking system based on multimodal fusion, characterized in that: A method for courseware review and duplication checking based on multimodal fusion is implemented as described in any one of claims 1-5, comprising a classification pre-review module, a keyword comparison module, a content comparison module and an online determination module, wherein the classification pre-review module is used to perform classification pre-review on the courseware to be uploaded, the keyword comparison module is used to compare the keywords with the pre-classified labels of the courseware, the content comparison module is used to split the courseware to be uploaded into multimodal data forms and compare them with the corresponding content in the courseware resource pool, and the online determination module is used to determine whether the courseware to be uploaded has passed the review.

Images