Teaching resource informatization arrangement and auditing method for online education platform

Abstract

The invention relates to the technical field of teaching resource information auditing, in particular to a teaching resource informatization arrangement and auditing method for an online education platform. The method comprises the steps of obtaining a historical text and a current auditing text in teaching resources; according to the anaphora relationship of the combined knowledge points in the sentences of the historical text and the knowledge point vocabularies in the sentences, screening real combined knowledge points; according to the traversal condition of combined vocabularies formed by knowledge point vocabularies in the historical text and knowledge point vocabularies in the current audit text when traversing the knowledge point hierarchical classification tree, obtaining traversal feature values of the combined vocabularies; according to the reference text in the knowledge point hierarchical classification tree, the combination correctness degree of the real combination knowledge points in the current audit text is obtained, and the current audit text is audited in combination with the occurrence situation of the real combination knowledge points in the current audit text in the knowledge point hierarchical classification tree, the correctness degree and the traversal feature values. According to the invention, the credibility of the auditing result of the teaching resource information is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of teaching resource information review, and particularly relates to a method for information collation and review of teaching resources on an online education platform. Background Art

[0002] When collating and reviewing the teaching resource information of an online education platform, traditional methods analyze the appropriateness of content based on the occurrence of simple words in different reference review texts, which may lead to confusion in the relationship between ordinary knowledge points and the application background of combined knowledge points or the connection between knowledge points, resulting in incorrect judgment of the degree of sentence problems, and further reducing the credibility of the review results of the teaching resource information on the education platform. Summary of the Invention

[0003] In order to solve the problem of low credibility of the review results when the existing methods review teaching resource information, the purpose of the present invention is to provide a method for information collation and review of teaching resources on an online education platform, and the specific technical solutions adopted are as follows: The present invention provides a method for information collation and review of teaching resources on an online education platform, and the method includes the following steps: Obtain the historical text and the current review text in the teaching resources of the online education platform; According to the reference relationship of the combined knowledge points in the sentences of the historical text and the distribution of the knowledge point words in the sentences, screen out the true combined knowledge points; according to the traversal situation of the combined words formed by the knowledge point words in the historical text and the knowledge point words in the current review text when traversing the knowledge point hierarchical classification tree, determine the corresponding words of each knowledge point word in the current review text, and obtain the traversal characteristic value of the combined words; the knowledge point hierarchical classification tree is constructed based on the knowledge points included in the historical text; comprehensively screen the reference text according to the occurrence situation of the corresponding words in the historical text and the traversal characteristic value; According to the reference text in the knowledge point hierarchical classification tree, obtain the correct combination degree of the true combined knowledge points in the current review text; according to the occurrence situation, combination correct degree and traversal characteristic value of the true combined knowledge points in the knowledge point hierarchical classification tree in the current review text, determine the problem degree described by the true combined knowledge points in the current review text; Review the current review text based on the problem degree.

[0004] Preferably, the screening of the true combined knowledge points according to the reference relationship of the combined knowledge points in the sentences of the historical text and the distribution of the knowledge point words in the sentences includes: For any combined knowledge point: Respectively count the number of occurrences of the corresponding reference of the any combined knowledge point in each historical text; Record the statements containing any of the combined knowledge points and the statements containing the references of the combined knowledge points as the first statements; Respectively count the proportion of the number of each type of knowledge point vocabulary other than any of the combined knowledge points in the first statements. If the proportion is greater than the preset proportion threshold, then regard the corresponding other knowledge point vocabulary as the associated knowledge points of any of the combined knowledge points; Respectively count the first occurrence times of any of the combined knowledge points in the historical texts, and the first quantity of the statements that simultaneously contain all the knowledge point vocabularies included in any of the combined knowledge points; Screen the true combined knowledge points according to the proportion, the first occurrence times, and the first quantity.

[0005] Preferably, the screening of the true combined knowledge points according to the proportion, the first occurrence times, and the first quantity includes: Respectively calculate the cumulative sum of the proportions of all the knowledge point vocabularies other than any of the combined knowledge points in the first statements, and the sum value of the first occurrence times and the first quantity; Take the normalized result of the product between the cumulative sum and the sum value as the possibility value that any of the combined knowledge points is a true knowledge point; If the possibility value is greater than the preset first possibility threshold, then determine any of the combined knowledge points as the true combined knowledge point.

[0006] Preferably, the determination of the reference vocabulary of each knowledge point vocabulary in the current review text according to the traversal situation of the combined vocabulary composed of the knowledge point vocabulary in the historical text and the knowledge point vocabulary in the current review text when traversing the knowledge point hierarchical classification tree includes: Respectively combine each knowledge point vocabulary in the current review text with each knowledge point vocabulary in the historical text to obtain combined vocabularies; For any combined vocabulary: Traverse upward along the knowledge point hierarchical classification tree for the two knowledge point vocabularies in any combined vocabulary through the depth-first algorithm, and obtain the traversal characteristic value of any combined vocabulary according to the traversal situation; For any knowledge point vocabulary in the current review text, regard the knowledge point vocabulary in the historical text in the combined vocabulary corresponding to the minimum traversal characteristic value in the combined vocabulary where any knowledge point vocabulary in the current review text is located as the reference vocabulary of any knowledge point vocabulary in the current review text.

[0007] Preferably, the obtaining of the traversal characteristic value of any combined vocabulary according to the traversal situation includes: Taking the sum of the traversal layers of the two knowledge points in any combined vocabulary when the traversed positions coincide as the traversal characteristic value of any combined vocabulary.

[0008] Preferably, screening reference texts according to the occurrence and traversal eigenvalue of the comprehensive comparison vocabulary in historical texts includes: For any knowledge point vocabulary in the current review text: If the traversal eigenvalue of the combined vocabulary formed by any knowledge point vocabulary in the current review text and its comparison vocabulary is less than or equal to a preset first threshold, then count the proportion of the first occurrence of the any knowledge point vocabulary in the historical text where its comparison vocabulary is located; Count the second quantity of the combined vocabulary whose traversal eigenvalue of all the combined vocabulary where the knowledge point vocabulary in the current review text is located is less than a preset second threshold; For any historical text, obtain the reference value of the any historical text to the current review text according to the sum of the second quantity, the proportion of the first occurrence of the knowledge point vocabulary in the current review text in the any historical text, and the average value of the traversal eigenvalues of all the combined vocabulary formed by the knowledge point vocabulary in the current review text and the knowledge point vocabulary in the any historical text; if the reference value is greater than a preset reference value threshold, then use the any historical text as a reference text.

[0009] Preferably, obtaining the correct combination degree of the true combined knowledge points in the current review text according to the reference texts in the knowledge point hierarchy classification tree includes: Obtain the triple corresponding to each sentence in the current review text; Obtain the common traversal position of all the knowledge point vocabulary in each sentence of the current review text, and classify the knowledge points in the next level of the common traversal position as the hierarchical aliases of each knowledge point vocabulary; Replace the knowledge point vocabulary in the triple with the corresponding hierarchical alias to obtain the replaced triple corresponding to each sentence in the current review text; Use the hierarchical aliases of the vocabulary of the ordinary knowledge points included in the true combined knowledge points in all texts as the characteristic representations of each true combined knowledge point; Count the second occurrence of the characteristic representation of each true combined knowledge point existing in each sentence of the current review text in the characteristic representations of the sentences in all reference texts, and the frequency of the replaced triple corresponding to each sentence in the current review text in the replaced triples corresponding to the sentences in all reference texts; Obtain the correct combination degree of the true combined knowledge points in the current review text according to the second occurrence and the frequency.

[0010] Preferably, determining the problem degree described by the true combined knowledge points in the current review text according to the occurrence, correct combination degree and traversal eigenvalue of the true combined knowledge points in the knowledge point hierarchy classification tree includes: Separate the other knowledge point vocabulary in the triple where the real combined knowledge point appears in the current review text except for the vocabulary of the real combined knowledge point, and record it as the candidate vocabulary; Place all candidate vocabularies into the knowledge point hierarchical classification tree for traversal, obtain the two candidate vocabularies corresponding to the maximum value of the traversal feature value, and obtain the first difference between the correct combination degrees of the real combined knowledge points where the two candidate vocabularies are located; For any real combined knowledge point in the current review text, obtain the problem degree described by the any real combined knowledge point in the current review text according to the traversal feature values corresponding to the two candidate vocabularies, the sum value of the correct combination degree corresponding to the triple where the any real combined knowledge point is located, and the corresponding first difference.

[0011] Preferably, obtaining the problem degree described by the any real combined knowledge point in the current review text according to the traversal feature values corresponding to the two candidate vocabularies, the sum value of the correct combination degree corresponding to the triple where the any real combined knowledge point is located, and the corresponding first difference includes: Calculate the first product of the first difference between the traversal feature values corresponding to the two candidate vocabularies and the correct combination degree of the real combined knowledge points where the two candidate vocabularies are located; Obtain the problem degree described by the any real combined knowledge point in the current review text according to the first product and the sum value of the correct combination degree corresponding to the triple where the any real combined knowledge point is located. The first product has a positive correlation with the problem degree, and the sum value of the correct combination degree has a negative correlation with the problem degree.

[0012] Preferably, the auditing of the current review text based on the problem degree includes: If the problem degree is greater than the preset problem degree threshold, determine the corresponding real combined knowledge point as the wrong combined knowledge point.

[0013] The present invention has at least the following beneficial effects: First, according to the corresponding results after anaphora resolution of the historical texts in the teaching resources of the online education platform in combination with knowledge points and the distribution of knowledge point vocabulary in the sentences, this invention screened out multiple true combined knowledge points. Then, a hierarchical classification tree of knowledge points was constructed based on the knowledge points contained in the historical texts. According to the traversal situation of the combined vocabulary formed by the knowledge point vocabulary in the historical texts and the knowledge point vocabulary in the current review text when traversing the hierarchical classification tree of knowledge points, reference texts were screened out, which have greater reference value for the current review text. Subsequently, the current review text will be analyzed based on the similarity between these reference texts and the current review text before, reducing the computational amount. Furthermore, the correct combination degree of the true combined knowledge points was evaluated, and the current review text was reviewed by comprehensively considering the occurrence situation and traversal eigenvalue of the true combined knowledge points in the current review text in the hierarchical classification tree of knowledge points, improving the credibility of the review results of the information of the teaching resources on the education platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0015] Figure 1 It is a flowchart of a method for information collation and review of teaching resources on an online education platform provided by an embodiment of the present invention; Figure 2 It is a structural block diagram of a system for information collation and review of teaching resources on an online education platform provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following will be described in detail with reference to the drawings and preferred embodiments for a method for information collation and review of teaching resources on an online education platform proposed by the present invention.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0018] The following will specifically describe the specific solution of a method for information collation and review of teaching resources on an online education platform provided by the present invention with reference to the drawings.

[0019] An embodiment of a method for information collation and review of teaching resources on an online education platform: The specific scenario targeted by this embodiment is as follows: The informatization arrangement and review of teaching resources is the core mechanism for ensuring the quality of educational platforms. By establishing a strict review system, errors, outdated content, and pseudoscience can be effectively screened out, ensuring the accuracy and credibility of knowledge dissemination. This embodiment will judge the combined knowledge points based on the lexical relationships in the text, and judge the problem degree of the combined knowledge points in the current review text by the similarity between the upper-level classification of the knowledge point vocabulary in the historical text and the current review text, so as to realize the review and arrangement of the current review text.

[0020] This embodiment proposes a method for the informatization arrangement and review of teaching resources on an online education platform, as Figure 1 shown, a method for the informatization arrangement and review of teaching resources on an online education platform in this embodiment includes the following steps: Step S1, obtain the historical text and the current review text in the teaching resources of the online education platform.

[0021] This embodiment first obtains all texts from the existing educational corpus, and records the obtained such texts as historical texts, and at the same time obtains the current review text.

[0022] Step S2, screen the true combined knowledge points according to the reference relationship of the combined knowledge points in the sentences of the historical text and the distribution of the knowledge point vocabulary in the sentences; determine the corresponding vocabulary of each knowledge point vocabulary in the current review text according to the traversal situation of the combined vocabulary composed of the knowledge point vocabulary in the historical text and the knowledge point vocabulary in the current review text when traversing the knowledge point hierarchical classification tree; the knowledge point hierarchical classification tree is constructed based on the knowledge points included in the historical text; comprehensively screen the reference text according to the occurrence situation of the corresponding vocabulary in the historical text and the traversal feature value.

[0023] There may be a situation in teaching resources where two different knowledge points are combined and described. The knowledge fields and application backgrounds etc. associated with the new knowledge point vocabulary generated after combination may be different from the original knowledge points. Therefore, when analyzing the appropriateness of the content by a simple vocabulary matching method for a certain combined knowledge point in the new teaching resources, it may cause confusion in the relationship between the ordinary knowledge points and the application background or related knowledge points of the combined knowledge points, resulting in misjudgment of the teaching content in the new teaching resources.

[0024] To accurately judge the combined knowledge points in teaching resources, the reference resolution of the vocabulary in a single resource text can be carried out. When a certain combination of knowledge points in the resource text is referred to more, this combination of knowledge points directly appears more in the sentences included in the resource text, and other related knowledge points appear more consistently, this combined knowledge point is more established.

[0025] Match and identify the common knowledge point vocabulary in these historical texts through the KMP algorithm. Obtain the reference relationships in the text of teaching resources through the pre-trained language model BERT. For example, in the sentence "Atrioventricular node ablation and pacemaker implantation as alternative therapies for reentrant tachycardia", the combined operation of atrioventricular node ablation and pacemaker implantation is referred to as this alternative therapy and used in other sentences.

[0026] Next, this embodiment will be described by taking any combined knowledge point in the historical text as an example. For other combined knowledge points in the historical text, the method provided in this embodiment can be used for processing.

[0027] Specifically, for any combined knowledge point: Count the number of occurrences of the corresponding reference of this combined knowledge point in each historical text respectively; record the sentences containing this combined knowledge point and the sentences containing the reference of this combined knowledge point as the first sentences; count the quantity proportion of each other knowledge point vocabulary except this combined knowledge point in all the first sentences respectively. The larger the quantity proportion, the more relevant the corresponding other knowledge point vocabulary is to this combined knowledge point. Therefore, if the quantity proportion is greater than the preset proportion threshold, the corresponding other knowledge point vocabulary is used as the associated knowledge point of this combined knowledge point; in this embodiment, the preset proportion threshold is 0.5, and in specific applications, the implementer can set it according to the specific situation.

[0028] The more times this combined knowledge point appears in the historical text, the more sentences that contain all the knowledge point vocabulary included in this combined knowledge point, and the greater the occurrence proportion of each larger associated knowledge point, the more established this combined knowledge point is. Therefore, count the number of times this combined knowledge point appears in the historical text, and record this number as the first number; count the number of sentences that contain all the knowledge point vocabulary included in this combined knowledge point, and record this number as the first quantity; calculate the cumulative sum of the quantity proportions of all other knowledge point vocabulary except this combined knowledge point in the first sentences respectively, and the sum value of the first number and the first quantity; take the normalized result of the product between this cumulative sum and this sum value as the possibility value that this combined knowledge point is a true knowledge point; if the possibility value is greater than the preset first possible threshold, then determine this combined knowledge point as a true combined knowledge point. In this embodiment, there are many data normalization methods, and an existing linear normalization method can be selected to normalize the data so that the value range of the obtained result after normalization is (0, 1). In this embodiment, the preset first possible threshold is 0.7, and in specific applications, the implementer can set it according to the specific situation.

[0029] Using the above method, multiple true knowledge points can be screened out from the historical text.

[0030] To reduce the excessive running consumption during text search and reference, it is necessary to roughly and simply screen the historical texts. The historical texts are simply screened based on the hierarchical interval between the vocabulary in the historical texts and the vocabulary in the current review text in the knowledge point hierarchical classification tree, as well as the presence performance of the similar classified vocabulary in the historical texts.

[0031] The knowledge point vocabulary contained in all historical texts is classified into a knowledge point hierarchical classification tree through the Classification and Regression (C&R) tree node algorithm. The process of establishing the knowledge point hierarchical classification tree is prior art and will not be elaborated here.

[0032] Each knowledge point vocabulary in the current review text is combined with each knowledge point vocabulary in the historical texts respectively to obtain multiple combined vocabularies. Each combined vocabulary is composed of one knowledge point vocabulary in the current review text and one knowledge point vocabulary in the historical texts.

[0033] For any combined vocabulary: Traverse the two knowledge point vocabularies in this combined vocabulary upward along the knowledge point hierarchical classification tree through the Depth-First Search (DFS) algorithm, and take the sum of the traversal levels of the two knowledge points in the combined vocabulary when the traversed positions coincide as the traversal eigenvalue of this combined vocabulary. By using this method, the traversal eigenvalue of each combined vocabulary can be obtained.

[0034] For any knowledge point vocabulary in the current review text, since there are multiple historical texts, this knowledge point vocabulary may exist in multiple combined vocabularies at the same time. Take the knowledge point vocabulary in the historical text in the combined vocabulary corresponding to the minimum traversal eigenvalue among the combined vocabularies where this knowledge point vocabulary is located as the reference vocabulary for this knowledge point vocabulary in the current review text. By using this method, the reference vocabulary for each knowledge point vocabulary in the current review text can be obtained.

[0035] For any knowledge point vocabulary in the current review text: If the traversal eigenvalue of the combined vocabulary composed of this knowledge point vocabulary in the current review text and its reference vocabulary is less than or equal to the preset first threshold, then count the proportion of the number of times this knowledge point vocabulary appears in the historical text where its reference vocabulary is located. By using the above method, the proportion of the number of times each knowledge point vocabulary in the current review text appears in the historical text where its reference vocabulary is located is counted respectively. In this embodiment, the preset first threshold is times the number of levels of the knowledge point hierarchical classification tree. In specific applications, the implementer can set it according to the specific situation.

[0036] Count the number of combined vocabularies whose traversal eigenvalues of all knowledge point vocabularies in the current review text are less than the preset second threshold, and record this number as the second number; in this embodiment, the preset second threshold is 5. In specific applications, the implementer can set it according to the specific situation.

[0037] For any historical text, based on the second quantity, the sum of the ratios of the number of occurrences of the knowledge point vocabulary in the current reviewed text in that historical text, and the average value of the traversal eigenvalues of all combined vocabulary formed by the knowledge point vocabulary in the current reviewed text and the knowledge point vocabulary in that historical text, the reference value of that historical text to the current reviewed text is obtained. The specific calculation formula for the reference value of the historical text to the current reviewed text is as follows: wherein, represents the reference value of the j-th historical text to the current reviewed text; represents the number of combined vocabulary whose traversal eigenvalues of all combined vocabulary where the knowledge point vocabulary in the current reviewed text is located are less than a preset second threshold, that is, the second quantity; represents the number of knowledge point vocabulary in the current reviewed text, represents the ratio of the number of occurrences of the i-th knowledge point vocabulary in the current reviewed text in the j-th historical text, represents the average value of the traversal eigenvalues of all combined vocabulary formed by the knowledge point vocabulary in the current reviewed text and the knowledge point vocabulary in the j-th historical text, is a normalization function.

[0038] The smaller the average value of the traversal eigenvalues of all combined vocabulary formed by the knowledge point vocabulary in the current reviewed text and the knowledge point vocabulary in the j-th historical text, and the larger the ratio of the number of occurrences of the i-th knowledge point vocabulary in the current reviewed text in the j-th historical text, when the sum of the traversal layers of the knowledge point vocabulary in the current reviewed text and the j-th historical text on the knowledge point hierarchical classification tree is smaller and the number of matching times is larger, it indicates that the content in the j-th historical text is similar to the content in the current reviewed text and has a relatively large proportion in the text. Therefore, the j-th historical text has a relatively high reference value for the current reviewed resource text.

[0039] By using the above method, the reference value of each historical text to the current reviewed text can be obtained. If the reference value is greater than the preset reference value threshold, the corresponding historical text is used as the reference text. In this embodiment, the preset reference value threshold is 0.7. In specific applications, the implementer can set it according to specific circumstances.

[0040] So far, in this embodiment, using the above method, multiple reference texts have been selected from all historical texts.

[0041] Step S3: Obtain the correct degree of combination of the actual combined knowledge points in the current reviewed text according to the reference text in the knowledge point hierarchical classification tree; determine the problem degree of the actual combined knowledge points described in the current reviewed text based on the occurrence situation, combination correct degree, and traversal eigenvalue of the actual combined knowledge points in the knowledge point hierarchical classification tree.

[0042] In the newly added resource text, there may be some combined knowledge points that do not appear in the reference text, but the knowledge points with the same combination using their upper-level classifications may exist in the reference text. When there are more such cases, this combined knowledge point is more likely to be correct.

[0043] For the common knowledge point vocabulary in each sentence of the current reviewed resource text, put multiple knowledge point vocabulary into the knowledge point hierarchical classification tree, and traverse through the depth-first search (DFS) algorithm to obtain the common traversal position of all knowledge point vocabulary. Use the knowledge point classification at the next level of this position as the classification alias for each knowledge point vocabulary. For example, in the sentence "Atrioventricular node ablation and pacemaker implantation as alternative therapies for reentrant tachycardia", the classifications of atrioventricular node ablation, pacemaker implantation, and alternative therapies are medical procedures, and the classification of reentrant tachycardia is disease.

[0044] Obtain triples for each sentence in each text through the GenerateIE algorithm. A triple consists of three elements, expressed as (subject, predicate / relationship, object), which is used to describe the association between entities. For example, "William Shakespeare created Hamlet", the subject is William Shakespeare, the predicate is created, and the object is Hamlet, so its corresponding triple is (William Shakespeare, created, Hamlet). Using this method, the triples corresponding to each sentence in the current reviewed text and the triples corresponding to each sentence in the reference reviewed text can be obtained.

[0045] Obtain the common traversal position of all knowledge point vocabulary in each sentence of the current reviewed text, and use the knowledge point classification at the next level of this common traversal position as the classification alias for each knowledge point vocabulary; replace the knowledge point vocabulary in the triple with the corresponding classification alias to obtain the replaced triple corresponding to each sentence in the current reviewed text.

[0046] Use the classification alias of the common knowledge point vocabulary included in the actual combined knowledge points in all texts as the feature representation of each actual combined knowledge point.

[0047] Count the number of times the feature representation of each true combined knowledge point in each statement of the current reviewed text appears in the feature representations of the statements in all reference texts. Denote this number as the second number. Each true combined knowledge point in each statement of the current reviewed text has a corresponding second number. Calculate the frequency of the replaced triples corresponding to each statement in the current reviewed text in the replaced triples corresponding to the statements in all reference texts. Next, based on the second number and the frequency, obtain the correct combination degree of the true combined knowledge points in the current reviewed text. Specifically, obtain the maximum value of all second numbers. For any true combined knowledge point in the current reviewed text, calculate the ratio of the second number corresponding to this true combined knowledge point to the maximum value of all second numbers, and take the product of this ratio and the frequency of the replaced triples corresponding to each statement in the current reviewed text in the replaced triples corresponding to the statements in all reference texts as the correct combination degree of this true combined knowledge point. The greater the correct combination degree, the more likely it is that the upper-level classification composition form of this true combined knowledge point meets the requirements and the content is correct. By using the above method, the correct combination degree of each true combined knowledge point in the current reviewed text can be obtained.

[0048] For any true combined knowledge point in the current reviewed text, respectively denote the other knowledge point vocabulary except the vocabulary of the true combined knowledge point in the triple where this true combined knowledge point is located as the candidate vocabulary; place all candidate vocabularies into the knowledge point hierarchical classification tree for traversal, obtain the two candidate vocabularies corresponding to the maximum traversal feature value, denote these two candidate vocabularies as the two candidate vocabularies corresponding to this true combined knowledge point, obtain the difference between the correct combination degrees of the true combined knowledge points where these two candidate vocabularies are located, and denote this difference as the first difference. The specific process of obtaining the difference between the correct combination degrees of the true combined knowledge points where these two candidate vocabularies are located is as follows: respectively calculate the average value of the correct combination degrees of the true combined knowledge points where each of these two candidate vocabularies is located, and take the absolute value of the difference between the average values of the correct combination degrees of the true combined knowledge points where these two candidate vocabularies are located as the difference between the correct combination degrees of the true combined knowledge points where these two candidate vocabularies are located. Then, calculate the product of the traversal feature values corresponding to the two candidate vocabularies corresponding to this true combined knowledge point and the first difference between the correct combination degrees of the true combined knowledge points where these two candidate vocabularies are located, and denote this product as the first product. Next, based on the first product and the sum value of the correct combination degree corresponding to the triple where this true combined knowledge point is located, obtain the problem degree described by this true combined knowledge point in the current reviewed text. The first product has a positive correlation with the problem degree, and the sum value of the correct combination degree has a negative correlation with the problem degree.

[0049] Among them, the positive correlation means that the dependent variable increases as the independent variable increases and decreases as the independent variable decreases, which can be an additive relationship, a multiplicative relationship, etc., and is determined by the actual application; the negative correlation means that the dependent variable decreases as the independent variable increases and increases as the independent variable decreases, which can be a subtractive relationship, a division relationship, etc., and is determined by the actual application.

[0050] In this embodiment, a specific calculation formula for the problem degree is given. The problem degree described by the k-th true combined knowledge point in the current review text can be expressed as: Among them, represents the problem degree described by the k-th true combined knowledge point in the current review text, represents the traversal eigenvalue corresponding to the two candidate words corresponding to the k-th true combined knowledge point, represents the first difference between the correct combination degrees of the true combined knowledge points where the two candidate words corresponding to the k-th true combined knowledge point are located, represents the correct combination degree corresponding to the i-th triple where the k-th true combined knowledge point is located, represents the number of triples where the k-th true combined knowledge point is located, represents a normalization function, so that the value range of the problem degree is (0, 1).

[0051] represents the first product, represents the sum value of the correct combination degrees corresponding to the triples where the k-th true combined knowledge point is located. When the traversal eigenvalues corresponding to the two candidate words corresponding to the k-th true combined knowledge point are larger, the difference between the correct combination degrees of the true combined knowledge points where the two candidate words corresponding to the k-th true combined knowledge point are located is larger, and the sum value of the correct combination degrees corresponding to the triples where the k-th true combined knowledge point is located is smaller, it indicates that the involved field of the k-th true combined knowledge point in the reviewed text is wider, and the difference in the correct combination degree is also larger, and the possibility of problems in the description in the text is greater, that is, the problem degree described by the k-th true combined knowledge point in the current review text is greater.

[0052] By using the above method, the problem degree described by each true combined knowledge point in the current review text can be obtained.

[0053] Step S4, review the current review text based on the problem degree.

[0054] In this embodiment, the problem degree described by each true combined knowledge point in the current review text is obtained in step S3. Next, the review of the current review text will be completed based on the problem degree.

[0055] Specifically, if the problem level is greater than the preset problem level threshold, the corresponding true combined knowledge points are determined as incorrect combined knowledge points, which can be directly marked and fed back to the submitter; if the problem level is less than or equal to the preset problem level threshold and cannot be matched in the historical text, the corresponding true combined knowledge points are transferred to manual review for further determination. The positions of the sentences with problems after review and the corresponding problem levels are stored correspondingly and transmitted to the database. The query statement of SQL is used to search for the problematic sentences, and the sentences with problems are fed back to the uploader for modification. In this embodiment, the preset problem level threshold is 0.7. In specific applications, the implementer can set it according to specific situations.

[0056] So far, the review of the current review text has been completed by using the method provided in this embodiment.

[0057] In this embodiment, first, according to the corresponding results after the anaphora resolution of the combined knowledge points in the sentences of the historical text in the teaching resources of the online education platform and the distribution of the knowledge point words in the sentences, multiple true combined knowledge points are screened out. Then, a knowledge point hierarchical classification tree is constructed based on the knowledge points included in the historical text. According to the traversal situation of the combined words formed by the knowledge point words in the historical text and the knowledge point words in the current review text when traversing the knowledge point hierarchical classification tree, reference texts are screened out. These texts have greater reference value for the current review text. Subsequently, the current review text will be analyzed based on the similarity between these reference texts and the current review text before, reducing the calculation amount. Furthermore, the correct combination degree of the true combined knowledge points is evaluated, and the current review text is reviewed by comprehensively considering the occurrence situation and traversal eigenvalue of the true combined knowledge points in the current review text in the knowledge point hierarchical classification tree, improving the credibility of the review result of the information of the teaching resources of the education platform.

[0058] An embodiment of an information-based sorting and review system for teaching resources of an online education platform: Refer to Figure 2 , which shows the structural block diagram of the information-based sorting and review system for teaching resources of an online education platform provided by an embodiment of the present invention. The system includes a data acquisition module, a screening module, an evaluation module, and a review module; Among them, the data acquisition module is used to acquire the historical text and the current review text in the teaching resources of the online education platform; A screening module, configured to screen out the truly knowledge-point-combined ones according to the reference relationships of knowledge points in the sentences of historical texts and the distribution of knowledge-point vocabulary in the sentences; determine the corresponding vocabulary of each knowledge-point vocabulary in the current review text according to the traversal situation of the combined vocabulary formed by the knowledge-point vocabulary in the historical text and the knowledge-point vocabulary in the current review text when traversing the knowledge-point hierarchical classification tree, and obtain the traversal feature value of the combined vocabulary; the knowledge-point hierarchical classification tree is constructed based on the knowledge points included in the historical text; screen the reference text by comprehensively considering the occurrence situation of the corresponding vocabulary in the historical text and the traversal feature value. An evaluation module, configured to obtain the correct combination degree of the truly knowledge-point-combined ones in the current review text according to the reference text in the knowledge-point hierarchical classification tree; determine the problem degree described by the truly knowledge-point-combined ones in the current review text according to the occurrence situation, combination correct degree and traversal feature value of the truly knowledge-point-combined ones in the knowledge-point hierarchical classification tree in the current review text. A review module, configured to review the current review text based on the problem degree.

[0059] It should be understood that Figure 2 The structural block diagram and modules of an informationization arrangement and review system for teaching resources of an online education platform shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented through hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated designed hardware. Those skilled in the art can understand that the above methods and systems can be implemented using computer-executable instructions and / or included in the processor control code. For example, such code is provided on a carrier medium such as a magnetic disk, CD, or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and modules in this specification can be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips and transistors, or programmable hardware devices such as field programmable gate arrays and programmable logic devices, but also by software executed by various types of processors, or by a combination of the above hardware circuits and software (for example, firmware).

[0060] For more details about the above various modules, reference can be made to other parts of this specification, and details will not be elaborated here.

[0061] In other embodiments, a medium is further provided, which stores at least one computer-executable program. When the at least one program is executed by a computer, the computer is caused to perform the steps in the method for information collation and review of teaching resources of the online education platform in the above embodiments. The medium may be a computer-readable storage medium.

[0062] Among them, the provided system and medium are both used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be elaborated here.

[0063] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for information-based collation and review of teaching resources on an online education platform, characterized in that, The method includes the following steps: Obtain the historical text and the current review text in the teaching resources of the online education platform; According to the reference relationship of the combined knowledge points in the sentences of the historical text and the distribution of the knowledge point vocabulary in the sentences, screen the true combined knowledge points; according to the traversal situation of the combined vocabulary composed of the knowledge point vocabulary in the historical text and the knowledge point vocabulary in the current review text when traversing the knowledge point hierarchical classification tree, determine the corresponding vocabulary of each knowledge point vocabulary in the current review text, and obtain the traversal characteristic value of the combined vocabulary; the knowledge point hierarchical classification tree is constructed based on the knowledge points included in the historical text; comprehensively consider the occurrence situation of the corresponding vocabulary in the historical text and the traversal characteristic value, and screen the reference text; According to the reference text in the knowledge point hierarchical classification tree, obtain the correct degree of combination of the true combined knowledge points in the current review text; according to the occurrence situation, combination correct degree and traversal characteristic value of the true combined knowledge points in the knowledge point hierarchical classification tree in the current review text, determine the problem degree described by the true combined knowledge points in the current review text; Review the current review text based on the problem degree.

2. The method for information-based collation and review of teaching resources of the online education platform according to claim 1, wherein The screening of the true combined knowledge points according to the reference relationship of the combined knowledge points in the sentences of the historical text and the distribution of the knowledge point vocabulary in the sentences includes: For any combined knowledge point: Respectively count the number of occurrences of the corresponding reference of any combined knowledge point in each historical text; Record the sentence containing any combined knowledge point and the sentence containing the reference of the combined knowledge point as the first sentence; Respectively count the proportion of the quantity of each other knowledge point vocabulary except any combined knowledge point in the first sentence. If the proportion is greater than the preset proportion threshold, then use the corresponding other knowledge point vocabulary as the associated knowledge point of any combined knowledge point; Respectively count the first number of occurrences of any combined knowledge point in the historical text and the first quantity of the sentences that simultaneously contain all the knowledge point vocabulary included in any combined knowledge point; Screen the true combined knowledge points according to the proportion, the first number and the first quantity.

3. The method for information-based collation and review of teaching resources of the online education platform according to claim 2, characterized in that, The screening of the true combined knowledge points according to the proportion, the first number and the first quantity includes: Respectively calculate the sum of the accumulations of the proportions of all other knowledge point vocabulary except any combined knowledge point in the first sentence and the sum value of the first number and the first quantity; use the normalized result of the product between the sum of the accumulations and the sum value as the possibility value of any combined knowledge point being a true knowledge point; If the possibility value is greater than the preset first possibility threshold, then determine any combined knowledge point as a true combined knowledge point.

4. The method for information collation and review of teaching resources of the online education platform according to claim 1, characterized in that, The determination of the corresponding vocabulary of each knowledge point vocabulary in the current review text according to the traversal situation of the combined vocabulary composed of the knowledge point vocabulary in the historical text and the knowledge point vocabulary in the current review text when traversing the knowledge point hierarchical classification tree includes: Respectively combine each knowledge point vocabulary in the current review text with each knowledge point vocabulary in the historical text to obtain combined vocabulary; For any combined vocabulary: Traverse upward along the knowledge point hierarchical classification tree for the two knowledge point vocabularies in the any combined vocabulary through a depth-first algorithm, and obtain the traversal eigenvalue of the any combined vocabulary according to the traversal situation; For any knowledge point vocabulary in the current review text, use the knowledge point vocabulary in the historical text of the combined vocabulary corresponding to the minimum traversal eigenvalue in the combined vocabulary where the any knowledge point vocabulary in the current review text is located as the comparison vocabulary of the any knowledge point vocabulary in the current review text.

5. The method for information-based collation and review of teaching resources of the online education platform according to claim 4, characterized in that, The obtaining the traversal eigenvalue of the any combined vocabulary according to the traversal situation includes: Taking the sum of the traversal layers of the two knowledge points in the any combined vocabulary when the traversed positions coincide as the traversal eigenvalue of the any combined vocabulary.

6. The method for information-based sorting and review of teaching resources of the online education platform according to claim 1, characterized in that, The screening of the reference text by comprehensively considering the occurrence situation of the comparison vocabulary in the historical text and the traversal eigenvalue includes: For any knowledge point vocabulary in the current review text: If the traversal eigenvalue of the combined vocabulary composed of the any knowledge point vocabulary in the current review text and its comparison vocabulary is less than or equal to a preset first threshold, then count the proportion of the first occurrence times of the any knowledge point vocabulary in the historical text where its comparison vocabulary is located; Count the second quantity of the combined vocabularies whose traversal eigenvalues of all the combined vocabularies where the knowledge point vocabularies in the current review text are located are less than a preset second threshold; For any historical text, obtain the reference value of the any historical text to the current review text according to the second quantity, the sum of the proportions of the first occurrence times of the knowledge point vocabularies in the current review text in the any historical text, and the average value of the traversal eigenvalues of all the combined vocabularies composed of the knowledge point vocabularies in the current review text and the knowledge point vocabularies in the any historical text; If the reference value is greater than a preset reference value threshold, then take the any historical text as the reference text.

7. The method for information-based collation and review of teaching resources of the online education platform according to claim 1, characterized in that, The obtaining the correct combination degree of the actual combined knowledge points in the current review text according to the reference text in the knowledge point hierarchical classification tree includes: Obtain the triple corresponding to each statement in the current review text; Obtain the common traversal position of all the knowledge point vocabularies in each statement in the current review text, and classify the knowledge points in the next level of the common traversal position as the hierarchical aliases of each knowledge point vocabulary; Replace the knowledge point vocabulary in the triple with the corresponding hierarchical alias to obtain the replaced triple corresponding to each statement in the current review text; Take the hierarchical alias of the vocabulary of the ordinary knowledge points included in the actual combined knowledge points in all the texts as the feature representation of each actual combined knowledge point; Count the second occurrence times of the feature representation of each actual combined knowledge point existing in each statement in the current review text in the feature representations of the statements in all the reference texts, and the frequency of the replaced triple corresponding to each statement in the current review text in the replaced triples corresponding to the statements in all the reference texts; Obtain the correct combination degree of the actual combined knowledge points in the current review text according to the second occurrence times and the frequency.

8. The teaching resource informatization collation and review method for the online education platform according to claim 1, characterized in that Determine the problem degree of the true combined knowledge point described in the current review text based on the occurrence, combination correctness, and traversal eigenvalue of the true combined knowledge point in the knowledge point hierarchical classification tree, including: Respectively record the other knowledge point vocabulary in the triple where the true combined knowledge point appears in the current review text except for the vocabulary of the true combined knowledge point as candidate vocabulary; Traverse all candidate vocabulary in the knowledge point hierarchical classification tree, obtain the two candidate vocabulary corresponding to the maximum traversal eigenvalue, and obtain the first difference between the combination correctness of the true combined knowledge points where the two candidate vocabulary are located; For any true combined knowledge point in the current review text, obtain the problem degree of the any true combined knowledge point described in the current review text according to the traversal eigenvalue corresponding to the two candidate vocabulary, the sum value of the combination correctness corresponding to the triple where the any true combined knowledge point is located, and the corresponding first difference; 9. The method for information-based sorting and review of teaching resources of the online education platform according to claim 8, characterized in that, The obtaining the problem degree of the any true combined knowledge point described in the current review text according to the traversal eigenvalue corresponding to the two candidate vocabulary, the sum value of the combination correctness corresponding to the triple where the any true combined knowledge point is located, and the corresponding first difference includes: Calculate the first product of the first difference between the traversal eigenvalue corresponding to the two candidate vocabulary and the combination correctness of the true combined knowledge points where the two candidate vocabulary are located; Obtain the problem degree of the any true combined knowledge point described in the current review text according to the first product and the sum value of the combination correctness corresponding to the triple where the any true combined knowledge point is located. The first product has a positive correlation with the problem degree, and the sum value of the combination correctness has a negative correlation with the problem degree; 10. The method for information-based collation and review of teaching resources of the online education platform according to claim 1, characterized in that, The auditing of the current review text based on the problem degree includes: If the problem degree is greater than the preset problem degree threshold, determine the corresponding true combined knowledge point as an incorrect combined knowledge point.

Images