Adaptive learning information recommendation method and device, equipment and storage medium

Abstract

The application provides a self-adaptive learning information recommendation method and device, equipment and a storage medium, wherein the method comprises the following steps: recommending a current learning resource to a learner based on a current learning strategy of the learner, and obtaining a current learning measurement result of the learner on the current learning resource; determining an adjustment parameter corresponding to the learner according to an expected learning target of the learner and the current learning measurement result; and adjusting the current learning strategy according to the expected learning target, the current learning measurement result and the adjustment parameter to obtain a target learning strategy of the learner. The adjustment parameter determined by the learning measurement result and the expected learning target can more accurately represent the difference between the learning condition of the learner and the learning target, thereby realizing timely adjustment of the current learning strategy to obtain a learning strategy that is more in line with the current learning condition of the learner, realizing dynamic recommendation of the learning resource and dynamic planning of the learning strategy, and thereby improving the accuracy of the learning resource recommendation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of adaptive learning, in particular to an adaptive learning information recommendation method and device, equipment and a storage medium. BACKGROUND

[0002] Adaptive learning is a learning method that formulates a learning solution suitable for the needs of learners by evaluating the individual characteristics of each stage of learners, and precisely customizes a learning plan for each learner.

[0003] In the existing adaptive learning system, the learning process, learning behavior, learning results and other factors of learners are generally recorded, the cognitive state of learners is automatically evaluated through learning analysis, and learning resources suitable for the individual characteristics of learners are recommended according to the characteristics of learners.

[0004] However, the current adaptive learning system is not accurate and flexible in formulating the learning path of learners, and therefore has the problem of low accuracy of learning resource recommendation. SUMMARY

[0005] The present application aims to solve the problem of low accuracy of learning resource recommendation in the prior art by providing an adaptive learning information recommendation method, device, equipment and storage medium.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present application is as follows:

[0007] In a first aspect, the present application provides an adaptive learning information recommendation method, which comprises:

[0008] recommending a current learning resource to the learner based on the current learning strategy of the learner, and obtaining a current learning measurement result of the learner on the current learning resource;

[0009] determining an adjustment parameter corresponding to the learner according to the expected learning goal of the learner and the current learning measurement result;

[0010] adjusting the current learning strategy according to the expected learning goal, the current learning measurement result and the adjustment parameter to obtain a target learning strategy of the learner, the target learning strategy being used to recommend a target learning resource to the learner, the matching degree between the target learning resource and the learner being greater than the matching degree between the current learning resource and the learner.

[0011] Optionally, the step of determining an adjustment parameter corresponding to the learner according to the expected learning goal of the learner and the current learning measurement result comprises:

[0012] inputting the historical measurement results of the current learning resources into a first deep neural network to obtain feature parameters of each of the current learning resources, the feature parameters comprising at least one of difficulty of the learning resources and distinguishability of the learning resources;

[0013] inputting the feature parameters of each of the current learning resources, the current measurement results and the current learning information of the learner into a second deep neural network to obtain adjustment parameters of the learner.

[0014] Optionally, before the inputting the feature parameters of each of the current learning resources, the current measurement results and the current learning information of the learner into the second deep neural network to obtain the adjustment parameters of the learner, the method further comprises:

[0015] identifying and storing current learning information of the learner in a process in which the learner learns each of the current learning resources, the current learning information comprising at least one of learning duration, learning times and currently learned learning resources.

[0016] Optionally, the adjusting the current learning strategy according to the expected learning target, the current learning measurement results and the adjustment parameters to obtain a target learning strategy of the learner comprises:

[0017] determining a learning bias of the learner according to the expected learning target and the current learning measurement results;

[0018] adjusting the current learning strategy according to the learning bias, a preset learning bias threshold and the adjustment parameters to obtain the target learning strategy of the learner.

[0019] Optionally, the adjusting the current learning strategy according to the learning bias, the preset learning bias threshold and the adjustment parameters to obtain the target learning strategy of the learner comprises:

[0020] if an absolute value of the learning bias is greater than the learning bias threshold, adding or reducing learning resources in the current learning strategy according to the adjustment parameters;

[0021] if the absolute value of the learning bias is less than or equal to the learning bias threshold, taking the current learning strategy as the target learning strategy of the learner.

[0022] Optionally, the method further comprises:

[0023] performing material labeling on learning materials to obtain labeling information of the learning materials, the labeling information comprising knowledge points to which the learning materials belong and difficulty of the learning materials;

[0024] The marking information of each learning material is identified by a data sensor, and each learning material is organized into the learning resource based on a self-organizing algorithm according to the marking information of each learning material.

[0025] Optionally, after the current learning strategy is adjusted according to the expected learning target, the current learning measurement result and the adjustment parameter to obtain the target learning strategy of the learner, the method further comprises:

[0026] The target learning resource is recommended to the learner based on the target learning strategy of the learner, and the target learning measurement result of the learner on the target learning resource is obtained.

[0027] The current learning information and the current measurement result of the learner on the current learning strategy and the target learning information and the target measurement result of the learner on the target learning strategy are combined and output to obtain the learning log of the learner.

[0028] In a second aspect, the present application provides an adaptive learning information recommendation device, the device comprising:

[0029] The recommendation module is configured to recommend a current learning resource to the learner based on the current learning strategy of the learner, and obtain a current learning measurement result of the learner on the current learning resource.

[0030] The determination module is configured to determine an adjustment parameter corresponding to the learner according to an expected learning target of the learner and the current learning measurement result.

[0031] The adjustment module is configured to adjust the current learning strategy according to the expected learning target, the current learning measurement result and the adjustment parameter to obtain a target learning strategy of the learner, the target learning strategy being used to recommend a target learning resource to the learner, the matching degree between the target learning resource and the learner being greater than the matching degree between the current learning resource and the learner.

[0032] Optionally, the determination module is specifically configured to:

[0033] The historical measurement result of the current learning resource is input into a first deep neural network to obtain a feature parameter of each current learning resource, the feature parameter comprising at least one of the following: difficulty of the learning resource, discrimination of the learning resource.

[0034] The feature parameter of each current learning resource, the current measurement result and the current learning information of the learner are input into a second deep neural network to obtain the adjustment parameter of the learner.

[0035] Optionally, the adjustment module is specifically configured to:

[0036] determine the learning bias of the learner according to the expected learning target and the current learning measurement result;

[0037] adjust the current learning strategy according to the learning bias, a preset learning bias threshold and the adjustment parameter, to obtain the target learning strategy of the learner.

[0038] Optionally, the adjustment module is further configured to:

[0039] if the absolute value of the learning bias is greater than the learning bias threshold, increase or decrease learning resources in the current learning strategy according to the adjustment parameter;

[0040] if the absolute value of the learning bias is less than or equal to the learning bias threshold, take the current learning strategy as the target learning strategy of the learner.

[0041] In a third aspect, the present application provides an electronic device, comprising: a processor, a storage medium and a bus, the storage medium stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine readable instructions to perform the steps of the method in the first aspect.

[0042] In a fourth aspect, the present application provides a computer readable storage medium, the computer readable storage medium stores a computer program, when the computer program is run by a processor, the steps of the method in the first aspect are executed.

[0043] The present application has the following beneficial effects: based on the current learning strategy of the learner, the current learning resources are recommended to the learner, and the current learning measurement result of the learner on the current learning resources is obtained, the adjustment parameter corresponding to the learner is determined according to the expected learning target and the current learning measurement result of the learner, and the current learning strategy is adjusted according to the expected learning target, the current learning measurement result and the adjustment parameter, to obtain the target learning strategy of the learner. The adjustment parameter determined by the learning measurement result and the expected learning target can more accurately represent the difference between the learning situation of the learner and the learning target, so as to realize the timely adjustment of the current learning strategy, to obtain a learning strategy more suitable for the current learning situation of the learner, realize the dynamic recommendation of learning resources and the dynamic planning of learning strategy, and thus improve the accuracy of learning resource recommendation. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those of ordinary skill in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0045] Figure 1 An architecture diagram of an adaptive learning system in the art is shown;

[0046] Figure 2 A flowchart of an adaptive learning information recommendation method provided by the embodiments of the present application is shown;

[0047] Figure 3 A flowchart of determining an adjustment parameter provided by the embodiments of the present application is shown;

[0048] Figure 4 A flowchart of determining a target learning strategy provided by the embodiments of the present application is shown;

[0049] Figure 5 A flowchart of determining a target learning strategy provided by the embodiments of the present application is shown;

[0050] Figure 6 A flowchart of obtaining a learning resource provided by the embodiments of the present application is shown;

[0051] Figure 7 A flowchart of outputting a learning log provided by the embodiments of the present application is shown;

[0052] Figure 8 A flowchart of another adaptive learning information recommendation method provided by the embodiments of the present application is shown;

[0053] Figure 9 A structural schematic diagram of an adaptive learning information recommendation device provided by the embodiments of the present application is shown;

[0054] Figure 10 A structural schematic diagram of an electronic device provided by the embodiments of the present application is shown. DETAILED DESCRIPTION

[0055] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of description and illustration, and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn according to the actual proportions. The flowchart used in the present application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can not be implemented in sequence, and the steps without logical context relationship can be reversed in sequence or implemented simultaneously. In addition, one or more other operations can be added to the flowchart or one or more operations can be removed from the flowchart under the guidance of the content of the present application.

[0056] In addition, the described embodiments are only some of the embodiments of the present application, not all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0057] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0058] In existing adaptive learning systems, the cognitive level of a learner is generally determined according to the answering situation of the learner, and then a static learning strategy is recommended for the learner.

[0059] However, first of all, the measurement result (i.e. the answering situation) of the learner is not a good consideration factor for learning content recommendation in the prior art, and since the learning cognitive level of the learner is implicit, it is difficult to directly measure the learning situation, and the static learning strategy recommendation in the prior art is difficult to adapt to the dynamically changing learning level of the learner. Therefore, the current adaptive learning system is not accurate and flexible in formulating the learning strategy of the learner, and there is a problem of low accuracy of learning resource recommendation.

[0060] Based on the above problems, the present application proposes an adaptive learning information recommendation method, which takes the measurement result of the learner as a reference factor for learning resource recommendation, and can realize dynamic and accurate planning of the learning path of the learner, such as Figure 1 As shown in FIG. 1, it is a schematic diagram of an adaptive learning system commonly used in the art.

[0061] Referring to FIG. 1, Figure 1The domain model defines the subject knowledge and concepts, providing content for the learner's learning. The student model represents the characteristics of the student, including personality characteristics, subject knowledge level, etc. The pedagogical model defines the pedagogical rules, i.e., how to access the knowledge and concepts in the domain model according to the characteristics of the student defined in the student model. The adaptive engine can create and update the entire software environment of the domain concepts and links, and the adaptive mechanism itself uses the information in the pedagogical model, the student model, and the domain model to select, annotate, and present the learning content to the learner in a personalized manner. The interface module defines the interaction between the user and the adaptive learning system, and the data used by the interface module can be used to infer the characteristics of the user, so as to update the learning object and verify the student model.

[0062] Figure 1 The running mechanism of the adaptive learning system can be summarized as follows: the system intelligently selects appropriate content from the domain model according to the evaluation results of the learner and recommends the content to the learner. The specific implementation process can be as follows: the system records the learning process, learning behavior, learning results, and the time and number of times used by the learner, automatically evaluates the cognitive state of the learner through learning analysis, and feeds back the evaluated cognitive state to the learning recommendation system. The system recommends learning resources suitable for the individual characteristics of the learner according to the characteristics of the learner, and continuously adjusts the learning strategy according to the immediate learning feedback, so as to realize personalized learning.

[0063] The method of the present application can be applied to Figure 1 the adaptive learning system as shown in the figure to realize dynamic planning of the learning strategy of the learner, and the execution subject of the method can be an electronic device.

[0064] Next, the adaptive learning information recommendation method of the present application will be further described in combination with Figure 2 The adaptive learning information recommendation method of the present application will be further described in combination with Figure 2 The method comprises the following steps:

[0065] S201: recommending a current learning resource to the learner based on the current learning strategy of the learner, and obtaining a current learning measurement result of the learner on the current learning resource.

[0066] Optionally, the learning strategy can be a learning plan specified for the learner. It should be understood that the learning plan can include learning and testing of multiple knowledge points, so one learning strategy can correspond to multiple learning resources.

[0067] Optionally, the learning resource can be a learning material group of a knowledge point obtained by combining relevant learning materials for the knowledge point by the electronic device. The learning materials can include, for example, at least one of a pre-class preparation video, a pre-class preparation document, a knowledge point video explanation, a post-class test question, and a post-class test question explanation, etc.

[0068] Optionally, the learning measurement can be a quantitative representation of the current learning resource reaching the expected learning goal submitted by the learner after learning the current learning resource, for example, a score form can be used as a learning measurement.

[0069] It should be noted that the learning measurement is a mathematical description and value determination of the learning behavior of the student, which can achieve the quantification of the degree of the student reaching the teaching goal. The learner and the learning environment form a dual relationship, and the implicit state or level of the learner can be indirectly measured by measuring the result of the interaction between the learner and the learning environment.

[0070] It should be noted that the learning strategy can be a learning plan for a course, and for a first-time user, the electronic device can use the preset default learning strategy of the course as the current learning strategy of the learner.

[0071] S202: determining the adjustment parameter corresponding to the learner according to the expected learning goal of the learner and the current learning measurement.

[0072] Optionally, the expected learning goal can be a preset value of the expected learning effect of the learner according to the teaching goal of the teacher or the past learning situation of the learner. For example, the teaching goal of the teacher for a knowledge point can be used as the expected learning goal of the knowledge point, or the teaching goal of the teacher for a course can be used as the expected learning goal of the course, or the expected learning goal can be determined by combining the teaching goal of the teacher and the past learning situation of the learner.

[0073] For example, assuming that the teaching goal of the teacher is to master knowledge point one, the expected learning goal can be to enable the learner to correctly complete the test questions corresponding to knowledge point one.

[0074] It should be noted that since the learner and the learning environment form a dual relationship, the learning situation of the learner can be measured by measuring the interaction result between the learner and the learning environment, for example, the current learning measurement of the learning resource.

[0075] Optionally, the expected learning goal of the learner and the current learning measurement can reflect the mastery of the learner to the knowledge point, and the electronic device can determine the adjustment parameter corresponding to the learner according to the mastery of the learner.

[0076] Optionally, the adjustment parameter can reflect the difference between the current learning situation of the learner and the expected learning goal, and the adjustment parameter can be introduced into the adaptive algorithm to adjust the learning path of the learner, so that the learning strategy is more suitable for the current learning situation of the learner. For example, the adaptive algorithm can be a graph neural network, an optimization algorithm, etc.

[0077] For example, the adjustment parameter can be represented by θ, which can be a numerical value (e.g., can be positive, negative or zero). When the adjustment parameter is positive, it can be understood that the recommended learning resource needs to be increased. When the adjustment parameter is negative, it can be understood that the recommended learning resource needs to be reduced. The smaller the absolute value of the numerical value, the smaller the degree of adjustment.

[0078] S203: Adjusting the current learning strategy according to the expected learning goal, the current learning measurement result and the adjustment parameter to obtain a target learning strategy of the learner. The target learning strategy is used to recommend a target learning resource to the learner. The matching degree between the target learning resource and the learner is greater than the matching degree between the current learning resource and the learner.

[0079] Optionally, the learning strategy can be adjusted, for example, the difficulty and quantity of each learning resource in the current learning strategy can be adjusted to obtain a learning strategy that is more suitable for the current learning situation of the learner.

[0080] Optionally, after determining the target learning strategy, the learning resources in the target learning strategy can be recommended to the learner as the target learning resource, so that the learner learns the target learning resource.

[0081] It should be noted that the knowledge level of the learner will change over time, and the demand for learning resources will also change accordingly. Therefore, after determining the target learning strategy, the above steps S201-S203 can be repeatedly executed to dynamically recommend learning resources to the learner and improve the accuracy of the learning strategy formulated for the learner.

[0082] Optionally, the matching degree between the learning resource and the learner can represent whether the learning strategy is suitable for the learner. It can be understood that if the matching degree between the learning resource and the learner is high, it means that the learning resource is more suitable for the current learning situation of the learner. In order to improve the accuracy of the recommended learning resource, the matching degree between the target learning resource and the learner should be greater than the matching degree between the current learning resource and the learner.

[0083] In the embodiments of the present application, the electronic device recommends the current learning resource to the learner based on the current learning strategy of the learner, and obtains the current learning measurement result of the learner on the current learning resource, determines the adjustment parameter corresponding to the learner according to the expected learning target of the learner and the current learning measurement result, adjusts the current learning strategy according to the expected learning target, the current learning measurement result and the adjustment parameter, and obtains the target learning strategy of the learner. The adjustment parameter determined in combination with the learning measurement result and the expected learning target can more accurately represent the difference between the learning situation of the learner and the learning target, so as to timely adjust the current learning strategy of the learner to obtain a learning strategy more suitable for the current learning situation of the learner, realize dynamic recommendation of learning resources and dynamic planning of learning strategies, and thus improve the accuracy of learning resource recommendation.

[0084] The following is a description of the step of determining the adjustment parameter corresponding to the learner according to the expected learning target of the learner and the current learning measurement result, as shown in the following table: Figure 3

[0085] S301: input the historical measurement result of the current learning resource into the first deep neural network to obtain the feature parameter of each current learning resource, and the feature parameter includes at least one of the difficulty of the learning resource and the discrimination degree of the learning resource.

[0086] Optionally, the historical measurement result of the current learning resource can be the answer record of the learner with different knowledge and ability levels for the current learning resource.

[0087] Optionally, the feature parameter of the learning resource can include the difficulty of the learning resource, the discrimination degree of the learning resource, etc.

[0088] As another possible implementation, the historical measurement result of all learning resources in the learning resource library can also be input into the first deep neural network before the above S201 step to obtain the feature parameter of each learning resource, and the current learning resource suitable for the current learning situation of the learner is determined from the learning resource library according to the feature parameter of each learning resource.

[0089] S302: input the feature parameter of each current learning resource, the current measurement result and the current learning information of the learner into the second deep neural network to obtain the adjustment parameter of the learner.

[0090] Optionally, inputting the feature parameter of each current learning resource, the current measurement result and the current learning information of the learner into the second deep neural network can determine the mastery of the learning resource with different difficulty and different discrimination degree by the learner, and determine the adjustment parameter of the learner.

[0091] ​Optionally, the learning information can be learning information recorded by the electronic device in the process of the learner learning the learning resource, for example, learning duration, learning times, and the like of the learner learning the learning resource.

[0092] For example, assuming that the current learning resource includes question 1, question 2, question 3, question 4, and question 5, and the difficulty and discrimination in the characteristic parameters thereof increase in turn, and the measurement result (taking correctness as the measurement result) of the learner A for each learning resource is correct, correct, correct, incorrect, and correct, respectively, then the measurement result, the characteristic parameters of each question, and the learning information of the learner learning each learning resource can be taken as the input of the second deep neural network, and the adjustment parameter for the learning condition of the learner at this time can be output by the second deep neural network.

[0093] It is worth noting that the first deep neural network and the second deep neural network can be a long short-term memory (LSTM) or a Transformer model.

[0094] In the embodiments of the present application, the characteristic parameters of the learning resource are obtained by the first deep neural network, and the adjustment parameter is determined by the second deep neural network according to the characteristic parameters of the learning resource, the current measurement result, and the current learning information of the learner, so that the numerical measurement of the learning level of the learner can be realized.

[0095] Before the characteristic parameters of each current learning resource, the current measurement result, and the current learning information of the learner are input into the second deep neural network to obtain the adjustment parameter of the learner, the learning information of the learner can be determined first, and the step S302 is further included before the step S302.

[0096] In the process of the learner learning each current learning resource, the current learning information of the learner is recognized and stored, and the current learning information includes at least one of the following: learning duration, learning times, and currently learned learning resource.

[0097] Optionally, in the process of the learner learning each current learning resource, the electronic device can recognize and record the learning information of the learner for each learning resource.

[0098] For example, assuming that the learning resource includes explanation video 1, explanation video 2, question 1, question 2, and explanation document 1, and in the process of the learner learning each learning resource, assuming that the learning resource already learned by the learner includes explanation video 1, explanation video 2, and question 1, the electronic device can record the learning times and learning duration of the learner for explanation video 1 and explanation video 2, and record the learning duration of the learner for question 1, and add an identifier to explanation video 1, explanation video 2, and question 1 to represent that the learning resource is a learned learning resource.

[0099] It is worth noting that in the embodiments of the present application, the learning level of the learner can be comprehensively judged in combination with the learning information of the learner, the learning measurement result and the expected learning target.

[0100] For example, assuming that the learning duration of the learner for the explanation video 1 is greater than the preset threshold, it can be explained that the learner does not firmly master the knowledge point corresponding to the explanation video 1, or there is difficulty in learning, and then the electronic device can recommend the learning resource with lower difficulty of the knowledge point for the learner in combination with the learning information.

[0101] For another example, assuming that the learning duration of the learner for the learning document 1 corresponding to the knowledge point 1 is much less than the preset threshold, but the learning duration of the learner for the explanation video 1 corresponding to the knowledge point 1 is within the normal range, then it can be judged that the learner prefers video explanation, and in the subsequent recommendation, the electronic device can preferentially recommend the explanation video of the knowledge point for the learner in combination with the learning information.

[0102] The following is a further description of the steps of adjusting the current learning strategy to obtain the target learning strategy of the learner according to the expected learning target, the current learning measurement result and the adjustment parameter, as shown in the following table. Figure 4 As shown in the following table, the step S203 includes:

[0103] S401: determining the learning deviation of the learner according to the expected learning target and the current learning measurement result.

[0104] Optionally, the learning deviation can be the difference between the expected learning target and the current learning measurement result, and the electronic device can first numerically value the expected learning target and the current learning measurement result, and then take the difference between the two as the learning deviation.

[0105] S402: adjusting the current learning strategy to obtain the target learning strategy of the learner according to the learning deviation, the preset learning deviation threshold and the adjustment parameter.

[0106] Optionally, the preset learning deviation threshold can be a constant threshold determined by the teacher according to the actual answer of the learner or determined by the electronic device according to the actual answer of all learners, and for example, the learning deviation threshold can be represented by σ, and in an ideal case, σ can be 0.

[0107] It is worth noting that according to the learning deviation and the preset learning deviation threshold, the electronic device can determine whether the current learning level of the learner reaches the expectation, and make corresponding adjustment to the current learning strategy.

[0108] The following is an explanation of the step of adjusting the learning strategy according to the learning deviation, the preset learning deviation threshold, and the adjustment parameter, to obtain the target learning strategy of the learner, as shown in the above step S402. Figure 5 The step S402 includes the following steps.

[0109] S501: If the absolute value of the learning deviation is greater than the learning deviation threshold, then the learning resources are increased or decreased in the current learning strategy according to the adjustment parameter.

[0110] Optionally, if the absolute value of the learning deviation is greater than the learning deviation threshold, i.e., the value of the expected learning target minus the learning measurement result is greater than σ or less than -σ, it can be indicated that the learner's mastery of the current knowledge point is not good, and the learning resources of the learner need to be increased or decreased.

[0111] Specifically, when the value of the expected learning target minus the learning measurement result is greater than σ, it can be indicated that the learner's mastery of the current knowledge point is not good. The learning resources of the learner need to be increased, and then at least one learning resource can be determined from the learning resource library according to the adjustment parameter, and these learning resources are added to the current learning strategy.

[0112] When the value of the expected learning target minus the learning measurement result is less than -σ, it can be indicated that the learner's mastery of the current knowledge point is far better than the preset target threshold, and the difficulty of the learning resources of the learner needs to be increased or some learning resources need to be reduced, so as to adjust the current learning strategy.

[0113] It should be noted that when the learning deviation is greater than the learning deviation threshold, the current learning resources may also be too difficult for the learner. Therefore, not only can the learning resources with difficulty more suitable for the learner be increased according to the adjustment parameter, but also the learning resources in the original learning strategy can be corrected, for example, the learning resources with difficulty greater than the current mastery level of the learner are moderately deleted, so that the adjusted learning strategy is more suitable for the learning level of the learner.

[0114] When the absolute value of the learning deviation is less than the learning deviation threshold, the current learning resources may also be too easy for the learner. Therefore, not only can the learning resources with difficulty lower than the mastery difficulty of the learner be reduced according to the adjustment parameter, but also at least one learning resource (for example, the learning resource with difficulty higher than the current mastery difficulty of the learner can be selected) can be determined from the learning resource library, and these learning resources are added to the current learning strategy, so that the adjusted learning strategy is more suitable for the learning level of the learner.

[0115] S502: If the absolute value of the learning deviation is less than or equal to the learning deviation threshold, then the current learning strategy is taken as the target learning strategy of the learner.

[0116] Optionally, if the absolute value of the learning deviation is less than or equal to the learning deviation threshold, that is, the value of the expected learning target minus the learning measurement result is less than or equal to σ, it can be indicated that the learner is good at the current knowledge point, and at this time, it can be considered that the best learning strategy and resource conforming to the current learning level of the learner are found, and therefore it is not necessary to adjust the learning resource, and the current learning strategy can be taken as the target learning strategy of the learner.

[0117] It is worth noting that taking a course as an example, a course can include multiple knowledge points, and the learning strategy of each knowledge point can include multiple groups of measurement questions (which can be understood as test questions) interspersed in the learning strategy. After the learner completes a group of measurement questions in a knowledge point and obtains a measurement result, the electronic device can adjust the subsequent learning strategy according to the adjustment parameter, the measurement result, and the expected target.

[0118] In the embodiments of the present application, the learning resource in the current learning strategy is adjusted according to the adjustment parameter, which can make the obtained learning strategy more suitable for the learning level of the learner, and realize the dynamic planning of the learning strategy of the learner.

[0119] As Figure 6 shown is a flowchart of obtaining a learning resource given by the present application, referring to Figure 6 , the steps of obtaining a learning resource include:

[0120] S601: The learning material is annotated to obtain annotation information of the learning material, and the annotation information includes: a knowledge point to which the learning material belongs, and a difficulty of the learning material.

[0121] Optionally, the learning material can be annotated by the first deep neural network in the above S301 step, for example, the learning material can be input into the first deep neural network to obtain the annotation information of the learning material. It should be understood that the annotation information can include the difficulty of the learning material, the discrimination degree of the learning material, and the knowledge point to which the learning material belongs, and it should be understood that these annotation information is only an example given by the present application, and the specific annotation information can be determined according to the actual implementation, which is not limited by the present application.

[0122] It is worth noting that the knowledge point to which the learning material belongs can be one or more, for example, in a comprehensive question, it can contain more than one knowledge point.

[0123] It should be noted that in order to improve the accuracy of the annotation of the learning material, the sample material annotated by the human can be input into the first deep neural network, and the first deep neural network is iteratively optimized according to the human-annotated information and the output result of the first deep neural network to improve the accuracy of the material annotation.

[0124] S602: Identify the label information of each learning material through the material sensor, and organize the learning materials into learning resources based on the self-organizing algorithm according to the label information of each learning material.

[0125] Optionally, the material sensor can be an algorithm for identifying the label information of the input learning material. After the labeling of all learning materials is completed in the above S601 step, the label information of each learning material can be identified through the material sensor, and the self-organizing algorithm is used to organize the learning materials into learning resources according to the label information.

[0126] The self-organizing algorithm is an unsupervised neural network algorithm that can classify input learning materials according to their label information.

[0127] As a possible implementation, learning materials belonging to the same knowledge point can be classified into a category based on the self-organizing algorithm and the knowledge graph, and learning materials belonging to the same knowledge point can be organized into learning resources for the knowledge point in order of difficulty and discrimination from low to high.

[0128] As another possible implementation, learning materials corresponding to multiple knowledge points belonging to the same course can be organized into learning resources for the course in order of difficulty and discrimination from low to high based on the self-organizing algorithm and the knowledge graph.

[0129] In the embodiments of the present application, by labeling the learning materials and organizing the learning materials into learning resources according to the label information, the available learning resources can be quickly determined when planning the learning strategy, and the efficiency of learning strategy planning is improved.

[0130] After the above adjustment of the current learning strategy according to the expected learning goal, the current learning measurement result and the adjustment parameter, the target learning strategy of the learner is obtained, as shown in Figure 7 After the above S203 step, the following steps are included:

[0131] S701: Recommend target learning resources to the learner based on the target learning strategy of the learner, and obtain the target learning measurement result of the learner on the target learning resources.

[0132] Optionally, after the target learning resources are recommended to the learner according to the target learning strategy, the learner can learn the target learning resources and submit the target learning measurement result of the target learning resources to the electronic device.

[0133] S702: Combine and output the current learning information and the current measurement result of the current learning strategy of the learner and the target learning information and the target measurement result of the target learning strategy, to obtain the learning log of the learner.

[0134] Optionally, the electronic device can combine the learning information corresponding to all the learning resources learned by the learner after starting learning and the measurement results into a learning log of the learner, and store the learning log in the database.

[0135] As another possible implementation, the electronic device can also combine the learning information corresponding to all the learning resources learned by the learner after starting learning into a learning log of the learner, and combine the measurement results of the learner into a score table of the learner, establish a mapping table between the learning log and the score table, and store the learning log, the score table and the mapping table in the database.

[0136] In the embodiments of the present application, by saving the learning information and measurement results of the learner as a learning log, the learner can know his / her learning situation at any time and consolidate learning in a timely manner according to the learning situation.

[0137] Next, the adaptive learning information recommendation method of the present application will be described in detail. Figure 8 The adaptive learning information recommendation method of the present application will be described in detail.

[0138] As shown in Figure 8 , the electronic device can first formulate a preliminary learning strategy according to the expected learning goal of the learner, and recommend the learning resources corresponding to the learning strategy to the learner by the execution mechanism according to the learning strategy. In the process of the learner learning the learning resources, the electronic device can identify and store the learning information of the learner, and obtain the output after learning of the learner (which can be understood as a learning log and a score table). According to the output after learning and the expected learning goal, the learning measurement result can be determined. According to the learning information and the output after learning, the electronic device can model the learner, i.e., determine the adjustment parameter. At the same time, the electronic device can also determine the learning deviation of the learner according to the learning measurement result and the learning goal, and finally adjust the learning strategy based on the adaptive algorithm (such as graph neural network, optimization algorithm, etc.) according to the adjustment parameter and the learning deviation.

[0139] Based on the same inventive concept, the embodiments of the present application also provide an adaptive learning information recommendation device corresponding to the adaptive learning information recommendation method. Since the principle of solving problems in the device of the embodiments of the present application is similar to the adaptive learning information recommendation method described above, the implementation of the device can be referred to the implementation of the method, and the repeated parts will not be described here.

[0140] Referring to Figure 9 , it is a schematic diagram of an adaptive learning information recommendation device provided by the embodiments of the present application. The device comprises a recommendation module 901, a determination module 902 and an adjustment module 903, wherein:

[0141] The recommendation module 901 is configured to recommend a current learning resource to the learner based on the current learning strategy of the learner, and to obtain a current learning measurement result of the learner on the current learning resource.

[0142] The determination module 902 is configured to determine an adjustment parameter of the learner according to the expected learning target and the current learning measurement result of the learner.

[0143] The adjustment module 903 is configured to adjust the current learning strategy according to the expected learning target, the current learning measurement result and the adjustment parameter, to obtain a target learning strategy of the learner, and the target learning strategy is used to recommend a target learning resource to the learner, and the matching degree between the target learning resource and the learner is greater than the matching degree between the current learning resource and the learner.

[0144] Optionally, the determination module 902 is specifically configured to:

[0145] input the historical measurement result of the current learning resource into a first deep neural network to obtain a feature parameter of each current learning resource, and the feature parameter includes at least one of the following: difficulty of the learning resource, and distinguishability of the learning resource;

[0146] input the feature parameter of each current learning resource, the current measurement result and the current learning information of the learner into a second deep neural network to obtain the adjustment parameter of the learner.

[0147] Optionally, the adjustment module 903 is specifically configured to:

[0148] determine a learning bias of the learner according to the expected learning target and the current learning measurement result;

[0149] adjust the current learning strategy according to the learning bias, a preset learning bias threshold and the adjustment parameter to obtain the target learning strategy of the learner.

[0150] Optionally, the adjustment module 903 is further specifically configured to:

[0151] if an absolute value of the learning bias is greater than the learning bias threshold, increase or decrease the learning resource in the current learning strategy according to the adjustment parameter;

[0152] if the absolute value of the learning bias is less than or equal to the learning bias threshold, the current learning strategy is taken as the target learning strategy of the learner.

[0153] Optionally, the device further comprises an identification module, and the identification module is specifically configured to:

[0154] identify and store current learning information of the learner in a process in which the learner learns each current learning resource, and the current learning information includes at least one of the following: learning duration, learning frequency and current learned learning resource.

[0155] Optionally, the apparatus further comprises an organization module, which is specifically configured to:

[0156] annotating the learning materials to obtain annotation information of the learning materials, the annotation information comprising: a knowledge point to which the learning materials belong, and a difficulty of the learning materials;

[0157] identifying the annotation information of each learning material through a material sensor, and organizing the learning materials into learning resources based on a self-organizing algorithm according to the annotation information of each learning material.

[0158] Optionally, the apparatus further comprises an output module, which is specifically configured to:

[0159] recommending a target learning resource to the learner based on the target learning strategy of the learner, and obtaining a target learning measurement result of the target learning resource by the learner;

[0160] combining and outputting the current learning information and the current measurement result of the current learning strategy of the learner, and the target learning information and the target measurement result of the target learning strategy, to obtain a learning log of the learner.

[0161] The description of the processing procedure of each module in the apparatus and the interaction procedure between the modules can refer to the related description in the above method embodiments, and will not be described in detail here.

[0162] The embodiments of the present application recommend a current learning resource to the learner based on the current learning strategy of the learner, and obtain a current learning measurement result of the current learning resource by the learner, determine an adjustment parameter corresponding to the learner according to an expected learning target and the current learning measurement result, adjust the current learning strategy according to the expected learning target, the current learning measurement result and the adjustment parameter, to obtain a target learning strategy of the learner. The adjustment parameter determined through the learning measurement result and the expected learning target can more accurately represent the difference between the learning situation of the learner and the learning target, so that the electronic device can timely adjust the current learning strategy of the learner to obtain a learning strategy more in line with the current learning situation of the learner, realize dynamic recommendation of learning resources and dynamic planning of learning strategies, and thus improve the accuracy of learning resource recommendation.

[0163] The embodiments of the present application also provide an electronic device, as shown in the accompanying drawings, which is a structural schematic diagram of the electronic device provided by the embodiments of the present application, comprising a processor 1001, a memory 1002 and a bus. The memory 1002 stores machine-readable instructions executable by the processor 1001 (such as a computer program, a computer program product, etc.). The processor 1001 executes the machine-readable instructions to perform the method embodiments of the present application. Figure 10 Figure 9 ​The processor 1001 and the memory 1002 are connected through a bus, and the processor 1001 executes the machine readable instructions to perform the adaptive learning information recommendation method when the computer device runs.

[0164] The computer readable storage medium stores a computer program, and the computer program is run by the processor to perform the steps of the adaptive learning information recommendation method.

[0165] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system and device can refer to the corresponding process in the method embodiment, and the present application will not be repeated here. In the several embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented by other ways. The above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed functional units can be indirect coupling or communication connection through some communication interface, device or module, and can be electrical, mechanical or other forms.

[0166] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. When the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or say the parts that make contributions to the prior art or parts of the technical solutions can be embodied in the form of software products, and the computer software products are stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0167] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. An adaptive learning information recommendation method, characterized in that, include: Based on the learner's current learning strategy, the system recommends current learning resources to the learner and obtains the learner's current learning measurement results for the current learning resources. Based on the learner's expected learning goals and the current learning measurement results, determine the adjustment parameters corresponding to the learner; Based on the expected learning objective, the current learning measurement results, and the adjustment parameters, the current learning strategy is adjusted to obtain the learner's target learning strategy. The target learning strategy is used to recommend target learning resources to the learner, and the matching degree between the target learning resources and the learner is greater than the matching degree between the current learning resources and the learner. The step of determining the adjustment parameters corresponding to the learner based on the learner's expected learning goals and the current learning measurement results includes: The historical measurement results of the current learning resource are input into the first deep neural network to obtain the feature parameters of each current learning resource. The feature parameters include at least one of the following: the difficulty of the learning resource and the discriminative power of the learning resource. The feature parameters of each current learning resource, the current learning measurement results, and the learner's current learning information are input into the second deep neural network to obtain the learner's adjustment parameters; The step of adjusting the current learning strategy based on the expected learning objective, the current learning measurement results, and the adjustment parameters to obtain the learner's target learning strategy includes: The learner's learning bias is determined based on the expected learning objectives and the current learning measurement results; Based on the learning deviation, the preset learning deviation threshold, and the adjustment parameters, the current learning strategy is adjusted to obtain the learner's target learning strategy.

2. The method according to claim 1, characterized in that, Before inputting the feature parameters of each of the current learning resources, the current learning measurement results, and the learner's current learning information into the second deep neural network to obtain the learner's adjustment parameters, the process further includes: During the process of the learner learning each of the current learning resources, the learner's current learning information is identified and stored. The current learning information includes at least one of the following: learning duration, number of learning attempts, and currently learned learning resources.

3. The method according to claim 1, characterized in that, The step of adjusting the current learning strategy based on the learning deviation, a preset learning deviation threshold, and the adjustment parameters to obtain the learner's target learning strategy includes: If the absolute value of the learning deviation is greater than the learning deviation threshold, then learning resources are increased or decreased in the current learning strategy according to the adjustment parameters. If the absolute value of the learning deviation is less than or equal to the learning deviation threshold, then the current learning strategy is taken as the learner's target learning strategy.

4. The method according to any one of claims 1-3, characterized in that, The method further includes: The learning materials are annotated to obtain the annotation information of the learning materials. The annotation information includes: the knowledge points to which the learning materials belong and the difficulty level of the learning materials. The annotation information of each learning material is identified by a data sensor, and the learning materials are organized into learning resources based on the annotation information of each learning material using a self-organizing algorithm.

5. The method according to any one of claims 1-3, characterized in that, After adjusting the current learning strategy based on the expected learning objective, the current learning measurement results, and the adjustment parameters to obtain the learner's target learning strategy, the method further includes: Based on the learner's goal-oriented learning strategy, target learning resources are recommended to the learner, and the learner's goal-oriented learning measurement results for the target learning resources are obtained. The learner's learning log is obtained by combining and outputting the learner's current learning information and current measurement results for the current learning strategy, as well as the target learning information and target measurement results for the target learning strategy.

6. An adaptive learning information recommendation device, characterized in that, include: The recommendation module is used to: recommend current learning resources to the learner based on the learner's current learning strategy, and obtain the learner's current learning measurement results for the current learning resources; The determining module is used to: determine the adjustment parameters corresponding to the learner based on the learner's expected learning objectives and the current learning measurement results; The adjustment module is used to: adjust the current learning strategy according to the expected learning objective, the current learning measurement result and the adjustment parameters to obtain the learner's target learning strategy. The target learning strategy is used to recommend target learning resources to the learner. The matching degree between the target learning resources and the learner is greater than the matching degree between the current learning resources and the learner. The determining module is used for: The historical measurement results of the current learning resource are input into the first deep neural network to obtain the feature parameters of each current learning resource. The feature parameters include at least one of the following: the difficulty of the learning resource and the discriminative power of the learning resource. The feature parameters of each current learning resource, the current learning measurement results, and the learner's current learning information are input into the second deep neural network to obtain the learner's adjustment parameters; The adjustment module is used for: The learner's learning bias is determined based on the expected learning objectives and the current learning measurement results; Based on the learning deviation, the preset learning deviation threshold, and the adjustment parameters, the current learning strategy is adjusted to obtain the learner's target learning strategy.

7. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus. The storage medium stores program instructions executable by the processor. When the electronic device is running, the processor communicates with the storage medium via the bus. The processor executes the program instructions to perform the steps of the adaptive learning information recommendation method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the adaptive learning information recommendation method as described in any one of claims 1 to 5.

Images