Digital resource data pushing method and system for intelligent teaching and medium
By obtaining and analyzing user data in smart teaching, performing similarity calculations based on content vectors and access behaviors, and screening out personalized recommendations, solving the problem of inaccurate push of digital resources in the existing technology, and achieving accurate and efficient digital resources Pushing improves teaching effect and user experience.
Patent Information
- Application Number
- CN202411394931.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-08
- Publication Date
- 2025-01-10
AI Technical Summary
The existing digital resource push methods have problems such as inaccurate, timely and unindividualized push in smart teaching, and cannot meet teaching needs.
By obtaining data in the digital resource library and user database, performing similarity calculations based on content vectors and access behavior data, filtering out a set of recommendation vectors, and recommending personalized digital resources according to the association rules.
It realizes accurate and efficient push of digital resources, improves teaching effect and user learning experience, and promotes the effective utilization and sharing of resources.
Smart Images
Figure CN119293329A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a method, system and medium for pushing digital resource data for smart teaching. Background Art
[0002] With the rapid development of information technology, smart teaching has become an important development direction in the field of education. In the process of smart teaching, the effective use and accurate push of digital resources are of great significance to improving teaching quality and efficiency. However, existing digital resource push methods often have problems such as inaccurate, untimely, and non-personalized push, which cannot meet the needs of smart teaching. Summary of the invention
[0003] To solve the above problems, the present invention provides a method, system and medium for pushing digital resource data for smart teaching, aiming to achieve accurate and efficient push of digital resources.
[0004] In order to achieve the above object, the present invention provides the following technical solutions: On the one hand, an embodiment of the present invention provides a method for pushing digital resource data for smart teaching, the method comprising the following steps: S100, obtaining content vectors corresponding to each digital resource in a digital resource library and access behavior data of each user in a user database; wherein the access behavior data includes a plurality of interaction data; the interaction data includes a plurality of keywords, an order of the plurality of keywords, and digital resources accessed by each keyword; S200, obtaining digital resources accessed by a target user, and matching and obtaining multiple keyword queues from the user database based on content tags of the digital resources; S300, performing similarity calculations on the plurality of keyword queues and the content vectors of each digital resource in the digital resource library, and filtering out a recommended vector set according to the calculated similarities; wherein the recommended vector set includes at least one content vector; S400: Recommending digital resources corresponding to each content vector in the recommendation vector set to the target user.
[0005] Optionally, the acquiring of the digital resources accessed by the target user, and obtaining multiple keyword queues by matching from the user database based on content tags of the digital resources, includes: S210, obtaining digital resources accessed by the target user; S220, marking m keywords included in the content tag of the digital resource as first keywords, and creating a corresponding keyword queue for each first keyword; S230, obtaining interaction data of each user in the user database, matching the first keyword with the keywords in each interaction data respectively, recording the interaction data with consistent matching as the first interaction data, and recording the remaining keywords in the first interaction data as the second keywords; S240, determining a correlation score of the second keyword based on the number of jumps between the second keyword and the first keyword; S250, determining the number n of the second keywords, if the number n of the second keywords is greater than 1, arranging the second keywords in descending order according to the corresponding relevance scores, and jumping to S260; if the number of the second keywords is 1, jumping to S270; S260, selecting a second keyword in sequence, marking it as the first keyword, and jumping to S230; S270, marking the second keyword as the first keyword, arranging each first keyword in the order of marking, numbering them, and adding them to the keyword queue to obtain m keyword queues.
[0006] Optionally, the relevance score of the second keyword k is calculated by the following formula: ; ; Wherein, T is the total number of first interaction data, t is the number of the first interaction data, K is the total number of second keywords, k is the number of the second keyword, is the number of jumps of the second keyword k in the t-th first interaction data, is the average jump probability of the second keyword k in the T first interaction data, and u is the association score of the second keyword k.
[0007] Optionally, the calculating similarities between the plurality of keyword queues and content vectors of each digital resource in the digital resource library, and selecting a recommended vector set according to the calculated similarities, includes: S310, record the first keyword in the keyword queue as Pij, take out the first keywords with the same number from the m keyword queues respectively, obtain Nj first keyword groups, form a first keyword vector Vj with the Nj first keyword groups, and obtain Nmax first keyword vectors; wherein, the first keyword group contains at least one first keyword, i represents the number of the keyword queue, i=1,2,...,m; j represents the number of the first keyword, j=1,2,...,Ni, Ni represents the total number of first keywords in the i-th keyword queue Pi, Nj represents the total number of first keywords with number j; Nmax represents the maximum value of the total number of first keywords Ni in the m keyword queues, Nmax=max(Ni); S320, for each first keyword vector, calculate the similarity between the first keyword vector and each content vector respectively, obtain the similarity between each content vector and the corresponding first keyword vector, select the content vector with a similarity greater than a similarity threshold as the recommended vector of the first keyword vector, and obtain a recommended vector set of Nmax first keyword vectors, wherein the recommended vector set includes at least one content vector.
[0008] Optionally, recommending the digital resources corresponding to each content vector in the recommendation vector set to the target user includes: S410, determining a recommendation score of the recommendation vector based on the similarity of the recommendation vector in the Nmax recommendation vector sets; S420, sorting the digital resources corresponding to the recommendation vectors in descending order of recommendation scores, and recommending them to the target user.
[0009] Optionally, the recommendation score of the recommendation vector is calculated by the following formula: ; in, is the recommendation score of the rth recommendation vector, is the similarity between the rth recommendation vector and the jth first keyword vector Vj; is the adjustment factor, 0< <1.
[0010] On the other hand, an embodiment of the present invention provides a digital resource data push system for smart teaching, including: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above method.
[0011] On the other hand, an embodiment of the present invention provides a computer-readable storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to perform the above method.
[0012] The beneficial effects of the present invention are as follows: the present invention discloses a method, system and medium for pushing digital resource data for smart teaching. After acquiring the digital resources accessed by the target user, the present invention matches multiple keyword queues from the user database based on the content tags of the digital resources, thereby mining keyword queues close to the target user from the user's interactive data; the multiple keyword queues are respectively calculated for similarity with the content vectors of each digital resource in the digital resource library, and a recommended vector set is screened out according to the calculated similarity, thereby screening out digital resources that meet the user's needs from the digital resource library and sending them to the user through the push system. The present invention recognizes different user groups, discovers the association between digital resources, and recommends relevant learning resources to users according to association rules; and realizes personalized and precise push of digital resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0014] Figure 1 It is a flow chart of a method for pushing digital resource data for smart teaching according to an embodiment of the present invention; Figure 2 It is a structural diagram of a digital resource data push system for smart teaching according to an embodiment of the present invention. DETAILED DESCRIPTION
[0015] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects disclosed in the present invention, so as to fully understand the purpose, scheme and effect disclosed in the present invention. It should be noted that the embodiments and features in the embodiments in this application can be combined with each other without conflict.
[0016] refer to Figure 1 ,like Figure 1 A method for pushing digital resource data for smart teaching provided by an embodiment of the present invention is shown, and the method comprises the following steps: S100, obtaining content vectors corresponding to each digital resource in a digital resource library and access behavior data of each user in a user database; wherein the access behavior data includes a plurality of interaction data; the interaction data includes a plurality of keywords, an order of the plurality of keywords, and digital resources accessed by each keyword; Specifically, the content vector includes a content ID, multiple keywords and a corresponding order, and each keyword is arranged in descending order according to the corresponding word frequency from large to small; when a user searches for a target digital resource by inputting a keyword and accesses the digital resource, an association between the digital resource and the keyword is established; the more users input the keyword, that is, the more times the keyword is searched when accessing the digital resource, the more the user recognizes the association between the keyword and the digital resource, and from the user's perspective, the greater the weight of the keyword.
[0017] When users browse digital resources, the browsing behavior of users is monitored to obtain interaction data; the interaction data includes the keywords entered by users when accessing the digital resource library, the digital resources accessed by the keywords, the order of keywords, etc.; by recording the interaction data of multiple users, a user database is formed.
[0018] S200, obtaining digital resources accessed by a target user, and matching and obtaining multiple keyword queues from the user database based on content tags of the digital resources; S300, performing similarity calculations on the plurality of keyword queues and the content vectors of each digital resource in the digital resource library, and filtering out a recommended vector set according to the calculated similarities; wherein the recommended vector set includes at least one content vector; S400: Recommending digital resources corresponding to each content vector in the recommendation vector set to the target user.
[0019] In the embodiment provided by the present invention, by collecting the user's interactive data, analyzing and mining using data mining technology, a personalized digital resource push strategy is generated, and digital resources that meet the user's needs are screened from the digital resource library and sent to the user through the push system. The present invention not only improves the teaching effect and user learning experience, but also promotes the effective use and sharing of digital resources.
[0020] The present invention collects, analyzes, mines and pushes teaching data, identifies different user groups, discovers the association between digital resources, and recommends relevant learning resources to users according to association rules; thus, the personalized and precise push of digital resources is achieved, thereby improving teaching effects and user learning experience.
[0021] In some embodiments, the acquiring of the digital resources accessed by the target user, and obtaining multiple keyword queues by matching from the user database based on the content tags of the digital resources, includes: S210, obtaining digital resources accessed by the target user; S220, marking m keywords included in the content tag of the digital resource as first keywords, and creating a corresponding keyword queue for each first keyword; S230, obtaining interaction data of each user in the user database, matching the first keyword with the keywords in each interaction data respectively, recording the interaction data with consistent matching as the first interaction data, and recording the remaining keywords in the first interaction data as the second keywords; S240, determining a correlation score of the second keyword based on the number of jumps between the second keyword and the first keyword; S250, determining the number n of the second keywords, if the number n of the second keywords is greater than 1, arranging the second keywords in descending order according to the corresponding relevance scores, and jumping to S260; if the number of the second keywords is 1, jumping to S270; S260, selecting a second keyword in sequence, marking it as the first keyword, and jumping to S230; S270, marking the second keyword as the first keyword, arranging each first keyword in the order of marking, numbering them, and adding them to the keyword queue to obtain m keyword queues.
[0022] It can be understood that m is a positive integer. By matching the first keyword with the keywords in each interaction data respectively, a second keyword that is highly correlated with the first keyword can be found from the interaction data. Then, the second keyword is used as a benchmark and marked as the first keyword, and the next second keyword with a high correlation is searched until the interaction data of each user in the user database is traversed, and m keyword queues are searched in order, preliminarily forming interaction data that is highly correlated with the digital resources accessed by the target user and the corresponding correlation strength, which provides a basis for the subsequent formation of a more accurate recommendation vector.
[0023] In some embodiments, the relevance score of the second keyword k is calculated by the following formula: ; ; Wherein, T is the total number of first interaction data, t is the number of the first interaction data, K is the total number of second keywords, k is the number of the second keyword, is the number of jumps of the second keyword k in the t-th first interaction data, is the average jump probability of the second keyword k in the T first interaction data, and u is the association score of the second keyword k.
[0024] It should be noted that the fewer the jump times, the stronger the correlation, the greater the average jump probability, and the higher the correlation score.
[0025] In some embodiments, the calculating similarities between the plurality of keyword queues and the content vectors of each digital resource in the digital resource library, and selecting a recommended vector set according to the calculated similarities, includes: S310, record the first keyword in the keyword queue as Pij, take out the first keywords with the same number from the m keyword queues respectively, obtain Nj first keyword groups, form a first keyword vector Vj with the Nj first keyword groups, and obtain Nmax first keyword vectors; wherein, the first keyword group contains at least one first keyword, i represents the number of the keyword queue, i=1,2,...,m; j represents the number of the first keyword, j=1,2,...,Ni, Ni represents the total number of first keywords in the i-th keyword queue Pi, Nj represents the total number of first keywords with number j; Nmax represents the maximum value of the total number of first keywords Ni in the m keyword queues, Nmax=max(Ni); S320, for each first keyword vector, calculate the similarity between the first keyword vector and each content vector respectively, obtain the similarity between each content vector and the corresponding first keyword vector, select the content vector with a similarity greater than a similarity threshold as the recommended vector of the first keyword vector, and obtain a recommended vector set of Nmax first keyword vectors, wherein the recommended vector set includes at least one content vector.
[0026] Specifically, first keywords with the same number indicate that the order of interaction is relatively consistent, and the formed first keyword vector can reflect the first keyword that is currently jumped to with a higher probability. The first keyword vector formed by multiple first keywords can more accurately reflect the content vector corresponding to the digital resource, thereby screening out the recommended vector set of the first keyword vector through the similarity between each content vector and the corresponding first keyword vector; the similarity between the content vector and the corresponding first keyword vector can be calculated using cosine similarity, and the similarity threshold value ranges from 0.2 to 0.8.
[0027] In some embodiments, recommending the digital resources corresponding to each content vector in the recommendation vector set to the target user includes: S410, determining a recommendation score of the recommendation vector based on the similarity of the recommendation vector in the Nmax recommendation vector sets; S420, sorting the digital resources corresponding to the recommendation vectors in descending order of recommendation scores, and recommending them to the target user.
[0028] Specifically, after the digital resources corresponding to the recommendation vectors are determined, the digital resources corresponding to the recommendation vectors in the recommendation vector set are recommended to the target user in order.
[0029] In some embodiments, the recommendation score of the recommendation vector is calculated by the following formula: ; in, is the recommendation score of the rth recommendation vector, is the similarity between the rth recommendation vector and the jth first keyword vector Vj; is the adjustment factor, 0< <1.
[0030] It should be noted that the similarity of the recommendation score of the recommendation vector is related to the number of times it appears. The more times it appears or the greater the similarity, the stronger the correlation of the recommendation vector and the higher the recommendation score; the obtained recommendation score has a higher accuracy.
[0031] refer to Figure 2 The embodiment of the present invention further provides a digital resource data push system for smart teaching, including: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above method.
[0032] The contents of the above method embodiments are all applicable to this embodiment. The functions specifically implemented by this embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments, which will not be repeated here.
[0033] Although the description of the present disclosure has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims by reference to the appended claims, taking into account the prior art, so as to effectively cover the intended scope of the present disclosure. In addition, the above description of the present disclosure is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present disclosure that have not yet been foreseen may still represent equivalent changes to the present disclosure.
Claims
1. A digital resource data push method for smart teaching, characterized in that: The method comprises the following steps: S100, obtaining content vectors corresponding to each digital resource in a digital resource library and access behavior data of each user in a user database; wherein the access behavior data includes a plurality of interaction data; the interaction data includes a plurality of keywords, an order of the plurality of keywords, and digital resources accessed by each keyword; S200, obtaining digital resources accessed by a target user, and matching and obtaining multiple keyword queues from the user database based on content tags of the digital resources; S300, performing similarity calculations on the plurality of keyword queues and the content vectors of each digital resource in the digital resource library, and filtering out a recommended vector set according to the calculated similarities; wherein the recommended vector set includes at least one content vector; S400: Recommending digital resources corresponding to each content vector in the recommendation vector set to the target user.
2. A method for pushing digital resource data for smart teaching according to claim 1, characterized in that: The step of acquiring the digital resources accessed by the target user and matching the digital resources to obtain multiple keyword queues from the user database based on the content tags of the digital resources includes: S210, obtaining digital resources accessed by the target user; S220, marking m keywords included in the content tag of the digital resource as first keywords, and creating a corresponding keyword queue for each first keyword; S230, obtaining interaction data of each user in the user database, matching the first keyword with the keywords in each interaction data respectively, recording the interaction data with consistent matching as the first interaction data, and recording the remaining keywords in the first interaction data as the second keywords; S240, determining a correlation score of the second keyword based on the number of jumps between the second keyword and the first keyword; S250, determining the number n of the second keywords, if the number n of the second keywords is greater than 1, arranging the second keywords in descending order according to the corresponding relevance scores, and jumping to S260; if the number of the second keywords is 1, jumping to S270; S260, selecting a second keyword in sequence, marking it as the first keyword, and jumping to S230; S270, marking the second keyword as the first keyword, arranging each first keyword in the order of marking, numbering them, and adding them to the keyword queue to obtain m keyword queues.
3. A method for pushing digital resource data for smart teaching according to claim 2, characterized in that: The relevance score of the second keyword k is calculated by the following formula: ； ； Wherein, T is the total number of first interaction data, t is the number of the first interaction data, K is the total number of second keywords, k is the number of the second keyword, is the number of jumps of the second keyword k in the t-th first interaction data, is the average jump probability of the second keyword k in the T first interaction data, and u is the association score of the second keyword k.
4. A method for pushing digital resource data for smart teaching according to claim 2, characterized in that: The step of calculating similarities between the plurality of keyword queues and content vectors of each digital resource in the digital resource library, and selecting a recommended vector set according to the calculated similarities, comprises: S310, record the first keyword in the keyword queue as Pij, take out the first keywords with the same number from the m keyword queues respectively, obtain Nj first keyword groups, form a first keyword vector Vj with the Nj first keyword groups, and obtain Nmax first keyword vectors; wherein, the first keyword group contains at least one first keyword, i represents the number of the keyword queue, i=1,2,...,m; j represents the number of the first keyword, j=1,2,...,Ni, Ni represents the total number of first keywords in the i-th keyword queue Pi, Nj represents the total number of first keywords with number j; Nmax represents the maximum value of the total number of first keywords Ni in the m keyword queues, Nmax=max(Ni); S320, for each first keyword vector, calculate the similarity between the first keyword vector and each content vector respectively, obtain the similarity between each content vector and the corresponding first keyword vector, select the content vector with a similarity greater than a similarity threshold as the recommended vector of the first keyword vector, and obtain a recommended vector set of Nmax first keyword vectors, wherein the recommended vector set includes at least one content vector.
5. A method for pushing digital resource data for smart teaching according to claim 4, characterized in that: The recommending the digital resources corresponding to each content vector in the recommendation vector set to the target user includes: S410, determining a recommendation score of the recommendation vector based on the similarity of the recommendation vector in the Nmax recommendation vector sets; S420, sorting the digital resources corresponding to the recommendation vectors in descending order of recommendation scores, and recommending them to the target user.
6. A method for pushing digital resource data for smart teaching according to claim 5, characterized in that: The recommendation score of the recommendation vector is calculated by the following formula: ； in, is the recommendation score of the rth recommendation vector, is the similarity between the rth recommendation vector and the jth first keyword vector Vj; is the adjustment factor, 0< <1.
7. A digital resource data push system for smart teaching, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 6.
8. A computer-readable storage medium storing a program executable by a processor, characterized in that: The processor-executable program is used to perform the method according to any one of claims 1 to 6 when executed by the processor.