Content pushing method and system based on intelligent teaching terminal

Abstract

The invention provides a content pushing method and system based on an intelligent teaching terminal, relates to the technical field of content pushing of the intelligent teaching terminal, and provides the content pushing method and system based on the intelligent teaching terminal. Multi-dimensional feature labeling is carried out on teaching resources in a teaching resource library, a resource knowledge graph is constructed, then feature extraction is carried out on learning behavior data of a current class, and extracted learning behavior features are converted into learning demand features needed by students of the current class through a preset semester demand rule library; teaching resources are matched from a teaching resource library on the basis of learning demand characteristics, and finally, push content lists which have different resource types and are matched with the learning demands of the students in the current class are screened out through expected utility evaluation and a resource diversity target optimization function, and are pushed to the intelligent teaching terminal. According to the method, the personalized learning requirements of the students are accurately captured, and the most suitable teaching resources are recommended according to the personalized learning requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of content push technology for smart teaching terminals, and in particular to a content push method and system based on a smart teaching terminal. Background Art

[0002] With the development of information technology and the popularization of smart teaching terminals, how to efficiently and accurately push teaching resources that suit students' learning needs has become an urgent problem. Traditional teaching resource push methods on smart teaching terminals often lack a deep understanding of students' individual needs. Different classes have different levels of class participation, knowledge mastery, and the difficulty level of teaching resources that suit the current class as a whole. Therefore, different classes require personalized and diverse resource push methods. In addition, traditional resource push algorithms are mostly based on keyword matching or simple user behavior analysis. Keyword-matching-based recommendation mechanisms mainly rely on the degree of match between keywords in resource titles, descriptions, or tags and user queries to recommend content. Although this method is simple and direct, it ignores the actual content quality and applicability of the resources, easily resulting in recommendations that are overly concentrated on a certain type of resource while ignoring resources that are more beneficial to specific student groups. This not only causes a lack of diversity in the types of resources pushed, but also leads to a poor match between teaching resources and student needs, affecting teaching effectiveness.

[0003] Therefore, it is necessary to provide a content push method and system based on a smart teaching terminal to solve the above technical problems. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a content push method and system based on a smart teaching terminal, which achieves the beneficial effect of personalized and diversified push of teaching resources to student groups in different classes.

[0005] The present invention provides a content push method based on a smart teaching terminal, the content push method comprising the following steps: S1: Build a resource knowledge graph based on the resources in the teaching resource library that have been annotated with multi-dimensional features, and match the teaching content on the current terminal with the resource knowledge graph to obtain the first candidate set; S2: Obtain the learning behavior data of the current class through the terminal and extract learning behavior features from the learning behavior data; S3: Using the preset learning demand rule library, the learning behavior characteristics are converted into learning demand characteristics with the same dimension as the resource characteristics annotated by the multi-dimensional features; S4: Based on the learning demand characteristics, the resources in the first candidate set are scored using a preset expected utility evaluation formula, the expected utility scores obtained are sorted in descending order, and at least two resources whose expected utility scores meet a preset scoring threshold are selected to form a second candidate set; S5: Filter out at least one resource from the second candidate set using a preset optimization algorithm, and combine them into a final push list to be pushed to the terminal.

[0006] Preferably, step S1 further includes presetting a set capacity upper limit threshold when acquiring the first candidate set.

[0007] Preferably, the step of obtaining the first candidate set is: Extracting a set of knowledge points contained in the current teaching content from the current terminal; And search the knowledge graph for resources related to the knowledge point set contained in the current teaching content, and combine them into candidate resources; The candidate resources are classified into a plurality of resource groups according to resource types, and resources are selected from the plurality of resource groups according to preset selection rules to form a first candidate set.

[0008] Preferably, the preset expected utility evaluation formula is: in, Score the expected utility, Score the relevance of requirements, Score your knowledge potential. Score engagement potential, 、 、 They are the fusion weights of demand relevance score, knowledge mastery potential score and participation potential score respectively.

[0009] Preferably, the demand relevance score is obtained by calculating the distance between the learning demand feature and the resource feature extracted from the resource that has been multi-dimensionally labeled.

[0010] Preferably, the calculation formula for the knowledge mastery potential score is: in, Score your knowledge potential. is the set of knowledge points contained in the current resource, It is the set of knowledge points that the class has mastered in the current resource. is the difficulty level of the current resource, Score the class's overall mastery of the knowledge set contained in the current resource.

[0011] Preferably, the participation potential score is calculated by obtaining the interaction type characteristics of the current resource and performing weighted fusion calculation based on a preset interaction type characteristic scoring standard.

[0012] Preferably, step S5 further includes using a preset resource diversity target optimization function to select resources, and the formula of the preset resource diversity target optimization function is: in, Optimize the function for resource diversity objective, For the The demand relevance score of each resource, The fusion weights for scoring the relevance of requirements, Score resource diversity, is the adjustment parameter.

[0013] Preferably, the resource diversity score is evaluated by calculating the proportion of different types of resources in the final push list.

[0014] The present invention provides a content push system based on a smart teaching terminal, which is applied to a content push method based on a smart teaching terminal. The content push system includes: A first candidate set generation module is used to construct a resource knowledge graph based on the resources in the teaching resource library that have been annotated with multi-dimensional features, and match the teaching content on the current terminal with the resource knowledge graph to obtain a first candidate set; A learning behavior feature extraction module is used to obtain the learning behavior data of the current class through the terminal and extract learning behavior features from the learning behavior data; A learning behavior feature conversion module is used to convert learning behavior features into learning demand features with the same dimension as resource features annotated with multi-dimensional features using a preset learning demand rule library; A second candidate set generation module is configured to score the resources in the first candidate set based on learning demand characteristics using a preset expected utility evaluation formula, sort the expected utility scores obtained by scoring in descending order, and select at least two resources whose expected utility scores meet a preset scoring threshold to form a second candidate set; The push list optimization module is used to select at least one resource from the second candidate set using a preset optimization algorithm and combine them into a final push list pushed to the terminal.

[0015] Compared with related technologies, the content push method based on the smart teaching terminal provided by the present invention has the following beneficial effects: The present invention provides a content push method based on a smart teaching terminal. It performs multi-dimensional feature annotation on the teaching resources in the teaching resource library and constructs a resource knowledge graph. It then extracts features from the learning behavior data of the current class and converts the extracted learning behavior features into learning demand features required by the students in the current class through a preset semester demand rule library. Based on the learning demand features, teaching resources are matched from the teaching resource library. Finally, a list of push content with different resource types that matches the learning needs of the students in the current class is screened out through the expected utility evaluation and resource diversity target optimization function, and pushed to the smart teaching terminal. The above push method can accurately capture the personalized learning needs of students and recommend the most appropriate teaching resources accordingly. In addition, the introduction of the expected utility evaluation formula comprehensively considers multiple dimensions such as demand relevance, knowledge mastery potential, and participation potential, ensuring the high quality and high value of the pushed resources. At the same time, the application of the resource diversity target optimization function further enriches the types and scope of pushed content, avoids the limitations of a single resource type, and helps stimulate students' interest and enthusiasm in learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of a content push method based on a smart teaching terminal of the present invention; Figure 2 This is a schematic diagram of the module structure of a content push system based on a smart teaching terminal of the present invention. DETAILED DESCRIPTION

[0017] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all of the structures. Furthermore, the embodiments of the present invention and the features of the embodiments may be combined with one another unless there is a conflict.

[0018] It should also be noted that, for ease of description, only portions relevant to the present invention are shown in the accompanying drawings, rather than all of the contents. Before discussing the exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the various operations (or steps) as being processed sequentially, many of the operations can be performed in parallel, concurrently, or simultaneously. In addition, the order of the various operations can be rearranged. The process can be terminated when its operations are completed, but may also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0019] Example 1 A content push method based on smart teaching terminal, in the specific implementation process, such as Figure 1 As shown, it shows a flow chart of a content push method based on a smart teaching terminal, and the content push method includes: Step S1: Construct a resource knowledge graph based on the resources in the teaching resource library that have been annotated with multi-dimensional features, and match the teaching content on the current terminal with the resource knowledge graph to obtain a first candidate set.

[0020] Specifically, step S1 further includes presetting a set capacity upper limit threshold when acquiring the first candidate set.

[0021] Specifically, the steps for obtaining the first candidate set are: Extracting a set of knowledge points contained in the current teaching content from the current terminal; And search the knowledge graph for resources related to the knowledge point set contained in the current teaching content, and combine them into candidate resources; The candidate resources are classified into a plurality of resource groups according to resource types, and resources are selected from the plurality of resource groups according to preset selection rules to form a first candidate set.

[0022] During the specific implementation process, the teaching resources in the teaching resource library are annotated with multi-dimensional features. The multi-dimensional features include but are not limited to knowledge points, difficulty levels, applicable grades, affiliated subjects, resource types, and interaction types. A structured resource knowledge graph is constructed for the resources in the teaching resource library that have been annotated with multi-dimensional features. The resource knowledge graph represents the relationship between each teaching resource and its relevance to different knowledge points, which facilitates subsequent matching with teaching content. After the resource knowledge graph is constructed, the knowledge point set contained in the ongoing teaching content is extracted from the current smart teaching terminal. Using the constructed resource knowledge graph, the knowledge point set contained in the extracted teaching content is matched with the resources in the resource knowledge graph to find resources associated with the knowledge point set contained in the teaching content. The filtered resources are combined into a first candidate set according to the preset filtering rules. A set capacity upper limit threshold is preset for the first candidate set. The setting of the set capacity upper limit threshold is based on the factor of the computational efficiency of the subsequent push content screening to avoid affecting the system response speed due to processing too much data. The acquisition of the first candidate set also needs to consider the diversity of resource types to ensure that the resources in the subsequent filtered push list are not single. First, the knowledge point set contained in the ongoing teaching activity is extracted from the current smart teaching terminal, and then all resources related to the knowledge point set contained in the ongoing teaching content are searched from the resource knowledge graph. The relevant resources found are combined to form a preliminary candidate resource pool. The preliminary candidate resource pool is classified according to resource type to form multiple resource groups of different resource types. Finally, according to the preset selection rules, a certain number of resources are selected from multiple resource groups and added to the first candidate set to ensure that the resources pushed to students are diverse and comprehensive, rather than limited to a certain form. The number of resources in the first candidate set is controlled based on the set capacity upper limit threshold to ensure the smooth progress of subsequent data processing.

[0023] Step S2: Acquire the learning behavior data of the current class through the terminal, and extract learning behavior features from the learning behavior data.

[0024] During the specific implementation process, the learning behavior data of the current class is collected from the teaching smart terminal. The learning behavior data includes but is not limited to the periodic assessment scores of knowledge points, the completion status of homework, and the participation data in the class question and answer session on the knowledge points on the current smart teaching terminal. After preprocessing the collected learning behavior data, the learning behavior characteristics are extracted. For example, the participation of all students in the class is calculated by summarizing the classroom interaction of all students, including but not limited to the frequency of answering questions, the proportion of participation in discussions, and the performance of completing group tasks. The periodic test scores and the statistical values ​​of the homework completion status of the students in the class are obtained through the academic affairs system to evaluate the overall mastery of the knowledge points by the current class students and the difficulty of receiving the current knowledge points.

[0025] Step S3: Using a preset learning requirement rule library, the learning behavior characteristics are converted into learning requirement characteristics with the same dimension as the resource characteristics annotated by the multi-dimensional characteristics.

[0026] During the specific implementation process, a detailed learning needs rule library is first constructed. The learning needs rule library is established based on the multi-faceted theoretical foundations of educational psychology, subject knowledge system and teaching experience. There are multiple mapping relationships, which convert learning behavior characteristics into learning needs characteristics. The content of the learning needs rule library may include but is not limited to defining the mastery of knowledge points at different levels and the overall difficulty of the current class students in accepting the knowledge points they are learning based on learning behavior data such as students' test scores and homework completion, thereby mapping them into the difficulty level of the corresponding pushed resources. The students' interest in the knowledge points they are learning is mapped according to their classroom interaction. The higher the student's interest, the more conducive it is to the mastery of the knowledge points and in-depth exploration. In the rule library, knowledge points with high student participation are mapped to resources with high difficulty levels and high interaction types for learning needs characteristics. Before conversion, in order to facilitate calculation, the learning behavior characteristics are converted into learning behavior feature vectors. When converting learning needs characteristics, considering the different effects of different features on learning effects, appropriate weight coefficients need to be assigned to each feature during the conversion process. Combining all the converted learning demand features, a comprehensive learning demand feature vector is formed. The learning demand feature vector reflects the characteristics of the push content required by the current learning status of the class, which facilitates subsequent feature matching in the teaching resource library. In addition, for better matching, the learning demand features need to be the same as the feature dimensions of the resources in the teaching resource library that have been multi-dimensionally labeled. This is achieved by mapping rules in the learning demand rule library to ensure that the learning demand features are directly mapped to specific resources in the resource library, thereby achieving accurate content recommendations.

[0027] Step S4: Based on the learning demand characteristics, use the preset expected utility evaluation formula to score the resources in the first candidate set, arrange the expected utility scores obtained by scoring in descending order, and select at least two resources whose expected utility scores meet the preset score threshold to form a second candidate set.

[0028] During the specific implementation process, based on the obtained learning demand characteristics, the expected utility score of the resources in the first candidate set is calculated using a preset expected utility evaluation formula. The expected utility score reflects the overall matching degree of the resources to the learning demand characteristics. The higher the expected utility score, the higher the matching degree of the resources to the learning demand characteristics, the more suitable it is for the learning situation of the current class, and it has a positive effect on promoting the students in the current class to the knowledge points they are learning. All resources in the first candidate set are sorted from high to low according to the expected utility score. According to the preset score threshold, at least two resources with expected utility scores that meet the requirements are selected and added to the second candidate set to ensure that there are enough choices in the final push list. The second candidate set also has a set capacity upper limit threshold to control the upper limit of the number of resources that can be accommodated in the second candidate set, reducing the amount of data processing calculations.

[0029] Specifically, the preset expected utility evaluation formula is: in, Score the expected utility, Score the relevance of requirements, Score your knowledge potential. Score engagement potential, 、 、 They are the fusion weights of demand relevance score, knowledge mastery potential score and participation potential score respectively.

[0030] During the specific implementation process, the expected utility evaluation formula includes three aspects: demand relevance, knowledge mastery potential score and participation potential score. The expected utility score is obtained through weighted fusion. The fusion weight is adjusted according to the teaching objectives and student characteristics. The expected utility score is used to screen resources to form the second candidate set.

[0031] Specifically, the demand relevance score is obtained by calculating the distance between the learning demand features and the resource features extracted from the resources that have been multi-dimensionally feature labeled.

[0032] During the specific implementation process, the demand relevance score represents the degree of relevance between the resource and the current learning demand. The learning demand characteristics are converted into a vector representation, and then the features of each resource in the first candidate set are extracted and converted into a resource feature vector. For example, for each resource in the first candidate set, the Euclidean distance between its resource feature vector and the current class learning demand feature vector is calculated as the demand relevance score of the resource.

[0033] Specifically, the calculation formula for the knowledge mastery potential score is: in, Score your knowledge potential. is the set of knowledge points contained in the current resource, It is the set of knowledge points that the class has mastered in the current resource. is the difficulty level of the current resource, Score the class's overall mastery of the knowledge set contained in the current resource.

[0034] During the specific implementation process, the knowledge mastery potential score represents the potential of resources to improve students' knowledge mastery. The knowledge mastery potential score formula takes into account the proportion of knowledge points mastered by the class in the current resource, the difficulty level of the resource, and the class's comprehensive mastery score of the set of knowledge points contained in the current resource. By considering multiple factors, it ensures that the pushed resources not only meet the overall learning needs of the students in the class, but also effectively improve knowledge mastery.

[0035] Specifically, the engagement potential score is calculated by obtaining the interaction type characteristics of the current resource and performing weighted fusion calculation based on the preset interaction type characteristic scoring criteria.

[0036] During implementation, we analyze the interaction characteristics of resources and pre-determine a scoring criteria for each interaction type. These criteria include, but are not limited to, frequency (the expected number of interactions); depth (whether the interactions delve into the topic, not just superficial responses); breadth (whether it attracts students of varying skill levels and the percentage of students participating in the class); and immediate feedback (whether timely feedback is provided to enhance learning outcomes). These scores are weighted and combined to produce an engagement potential score. This approach allows for a more accurate prediction of the impact of resources on student engagement, thereby optimizing recommendation strategies and improving teaching effectiveness.

[0037] Step S5: Filter out at least one resource from the second candidate set using a preset optimization algorithm, and combine them into a final push list to be pushed to the terminal.

[0038] Specifically, step S5 also includes using a preset resource diversity target optimization function to select resources. The formula of the preset resource diversity target optimization function is: in, Optimize the function for resource diversity objective, For the The demand relevance score of each resource, The fusion weights for scoring the relevance of requirements, Score resource diversity, is the adjustment parameter.

[0039] Specifically, the resource diversity score evaluates the diversity of the final push list by calculating the proportion of different types of resources.

[0040] During the specific implementation process, the resources in the final push list are screened from the second candidate set. In order to screen out resources with diverse resource types and ensure that the pushed content is not limited to a single type of content, a genetic algorithm is used to find the resource combination that maximizes the value of the resource diversity target optimization function. The screened resource combinations are combined into the final push list pushed to the terminal. In the generated final push list, the number of each type of resource is counted separately, and the proportion of each type of resource in the total number of resources is calculated. Finally, the concept of entropy is used to calculate resource diversity. Through the preset optimization algorithm and combined with the resource diversity target optimization function, it is ensured that the selected resources not only have a high expected utility score but also have a diversity of resource types.

[0041] The working principle of the content push method based on the smart teaching terminal provided by the present invention is as follows: First, the resources in the teaching resource library are annotated in detail through multi-dimensional feature annotation, and a structured resource knowledge graph is constructed based on this. A preliminary screening is performed to obtain the first candidate set, which provides a basis for subsequent resource matching and recommendation. Next, the class learning behavior data is collected and analyzed to extract learning behavior characteristics. Then, using the preset learning demand rule library, the learning behavior characteristics are converted into learning demand characteristics with the same feature dimensions as the resources in the teaching resource library that have been multi-dimensionally annotated, so as to directly match them with the resources in the resource knowledge graph. This process ensures that students' learning behaviors can be accurately mapped to learning needs and matched to specific educational resources. Secondly, based on the learning demand characteristics, the candidate resources are scored using the expected utility evaluation formula, sorted in descending order according to the expected utility score, and at least two resources whose expected utility scores meet the preset score threshold are selected to form the second candidate set. The expected utility evaluation formula comprehensively considers multiple dimensions such as demand relevance, knowledge mastery potential, and participation potential to comprehensively evaluate the value of each resource to the class student group. Finally, By applying the preset optimization algorithm and combining it with the resource diversity target optimization function, the optimal combination is screened out from the candidate resources. The optimal combination is the resource combination that maximizes the value of the resource diversity target optimization function. This resource combination is used as the final resource list pushed to the terminal, achieving the purpose of personalized and diversified teaching resource push for student groups in different classes.

[0042] Example 2 A content push system based on a smart teaching terminal is applied to a content push method based on a smart teaching terminal. In the specific implementation process, Figure 2 As shown, it shows a module structure diagram of a content push system based on a smart teaching terminal, and the content push system includes: The first candidate set generation module 100 is used to construct a resource knowledge graph based on the resources in the teaching resource library that have been annotated with multi-dimensional features, and match the teaching content on the current terminal with the resource knowledge graph to obtain a first candidate set; The learning behavior feature extraction module 200 is used to obtain the learning behavior data of the current class through the terminal and extract the learning behavior features from the learning behavior data; The learning behavior feature conversion module 300 is used to convert the learning behavior feature into a learning demand feature with the same dimension as the resource feature dimension annotated by the multi-dimensional feature using a preset learning demand rule library; A second candidate set generating module 400 is configured to score the resources in the first candidate set based on learning requirement characteristics using a preset expected utility evaluation formula, sort the expected utility scores obtained by scoring in descending order, and select at least two resources whose expected utility scores meet a preset scoring threshold to form a second candidate set; The push list optimization module 500 is configured to select at least one resource from the second candidate set using a preset optimization algorithm and combine the resources into a final push list to be pushed to the terminal.

[0043] The working principle of the content push system based on the smart teaching terminal provided by the present invention is as follows: First, the first candidate set generation module 100 constructs a resource knowledge graph based on the teaching resource library that has been annotated with multi-dimensional features, and matches the teaching content on the current terminal with it to generate the first candidate set. Next, the learning behavior feature extraction module 200 collects the learning behavior data of the current class through the terminal and extracts key learning behavior features from it. Then, the learning behavior feature conversion module 300 uses the preset learning demand rule library to convert these learning behavior features into learning demand features with the same dimension as the resource features to facilitate the subsequent matching process. The second candidate set generation module 400 uses the expected utility evaluation formula based on the learning demand features to score the resources in the first candidate set, and selects resources with expected utility scores higher than the preset threshold to form the second candidate set. Finally, the push list optimization module 500 applies the optimization algorithm to filter out at least one resource from the second candidate set, ensuring that the selected resources not only have a high expected utility score but also maintain a certain degree of diversity, and are combined into the resource list that is finally pushed to the terminal.

[0044] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0045] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium, including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, magnetic disk storage, or magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0046] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

Claims

1. A content push method based on a smart teaching terminal, characterized in that: The content push method comprises the following steps: S1: Build a resource knowledge graph based on the resources in the teaching resource library that have been annotated with multi-dimensional features, and match the teaching content on the current terminal with the resource knowledge graph to obtain the first candidate set; S2: Obtain the learning behavior data of the current class through the terminal and extract learning behavior features from the learning behavior data; S3: Using the preset learning demand rule library, the learning behavior characteristics are converted into learning demand characteristics with the same dimension as the resource characteristics annotated by the multi-dimensional features; S4: Based on the learning demand characteristics, the resources in the first candidate set are scored using a preset expected utility evaluation formula, the expected utility scores obtained are sorted in descending order, and at least two resources whose expected utility scores meet a preset scoring threshold are selected to form a second candidate set; S5: Filter out at least one resource from the second candidate set using a preset optimization algorithm, and combine them into a final push list to be pushed to the terminal.

2. A content push method based on a smart teaching terminal according to claim 1, characterized in that: Step S1 also includes presetting a set capacity upper limit threshold when obtaining the first candidate set.

3. A content push method based on a smart teaching terminal according to claim 2, characterized in that: The steps of obtaining the first candidate set are: Extracting a set of knowledge points contained in the current teaching content from the current terminal; And search the knowledge graph for resources related to the knowledge point set contained in the current teaching content, and combine them into candidate resources; The candidate resources are classified into a plurality of resource groups according to resource types, and resources are selected from the plurality of resource groups according to preset selection rules to form a first candidate set.

4. A content push method based on a smart teaching terminal according to claim 3, characterized in that: The preset expected utility evaluation formula is: in, Score the expected utility, Score the relevance of requirements, Score your knowledge potential. Score engagement potential, 、 、 They are the fusion weights of demand relevance score, knowledge mastery potential score and participation potential score respectively.

5. A content push method based on a smart teaching terminal according to claim 4, characterized in that: The demand relevance score is obtained by calculating the distance between the learning demand feature and the resource feature extracted from the resource that has been multi-dimensionally labeled.

6. A content push method based on a smart teaching terminal according to claim 5, characterized in that: The calculation formula for the knowledge mastery potential score is: in, Score your knowledge potential. is the set of knowledge points contained in the current resource, It is the set of knowledge points that the class has mastered in the current resource. is the difficulty level of the current resource, Score the class's overall mastery of the knowledge set contained in the current resource.

7. A content push method based on a smart teaching terminal according to claim 6, characterized in that: The participation potential score is calculated by obtaining the interaction type characteristics of the current resource and performing weighted fusion calculation based on a preset interaction type characteristic scoring standard.

8. The content push method based on the smart teaching terminal according to claim 7 is characterized in that: Step S5 also includes using a preset resource diversity target optimization function to select resources. The formula of the preset resource diversity target optimization function is: in, Optimize the function for resource diversity objective, For the The demand relevance score of each resource, The fusion weights for scoring the relevance of requirements, Score resource diversity, is the adjustment parameter.

9. The content push method based on the smart teaching terminal according to claim 8 is characterized in that: The resource diversity score evaluates the diversity of the final push list by calculating the proportion of different types of resources in the list.

10. A content push system based on a smart teaching terminal, characterized in that: Applicable to a content push method based on a smart teaching terminal as described in claims 1 to 9, the content push system includes: A first candidate set generation module is used to construct a resource knowledge graph based on the resources in the teaching resource library that have been annotated with multi-dimensional features, and match the teaching content on the current terminal with the resource knowledge graph to obtain a first candidate set; A learning behavior feature extraction module is used to obtain the learning behavior data of the current class through the terminal and extract learning behavior features from the learning behavior data; A learning behavior feature conversion module is used to convert learning behavior features into learning demand features with the same dimension as resource features annotated with multi-dimensional features using a preset learning demand rule library; A second candidate set generation module is configured to score the resources in the first candidate set based on learning demand characteristics using a preset expected utility evaluation formula, sort the expected utility scores obtained by scoring in descending order, and select at least two resources whose expected utility scores meet a preset scoring threshold to form a second candidate set; The push list optimization module is used to select at least one resource from the second candidate set using a preset optimization algorithm and combine them into a final push list pushed to the terminal.

Images