CN117274005B

CN117274005B - Big data pushing method and system based on digital education

Info

Publication number: CN117274005B
Application number: CN202311551986.7A
Authority: CN
Inventors: 易莉
Original assignee: Xichang College
Current assignee: Xichang College
Priority date: 2023-11-21
Filing date: 2023-11-21
Publication date: 2024-02-09
Anticipated expiration: 2043-11-21
Also published as: CN117274005A

Abstract

The utility model provides a big data pushing method and system based on digital education, which is characterized in that the prior online learning event of a target digital education user at a preset online learning node and the first learning mode data tracked at the preset online learning node are subjected to correlation analysis to generate event correlation analysis data, so that feature speculation and correlation configuration are further carried out, and the learning behaviors and learning modes of students can be accurately identified and speculated, so that high-quality teaching suggestions are generated. In addition, the priori learning events and learning modes of students are considered, individuation can be carried out according to specific requirements of each student, and further, in the process of carrying out big data pushing on the correlation configuration data, the optimization of a digital education pushing strategy is facilitated. The learning behavior and learning mode of the students can be dynamically analyzed, and the prediction and configuration can be carried out according to the trusted value, so that the students can dynamically adjust along with the learning process of the students, and the most suitable teaching suggestion can be provided.

Description

Big data pushing method and system based on digital education

Technical Field

The application relates to the technical field of digital education, in particular to a big data pushing method and system based on digital education.

Background

With the rapid development of internet technology, digital education becomes an important component of the education field. The online learning platform can provide a convenient and flexible learning mode, so that students can learn according to own rhythm and requirements. However, how to provide personalized teaching support for each student's characteristics becomes an important issue facing current digital education, as each student's learning habit, background knowledge, and understanding ability are different.

Currently, most online learning platforms generate teaching suggestions mainly by collecting and analyzing learning behavior data of students, such as learning time, learning progress, answering situations, and the like. However, this method often cannot accurately reflect the learning mode of the student, and cannot fully utilize the prior learning event information. Moreover, these systems often fail to provide accurate, comprehensive teaching advice due to the lack of efficient correlation analysis and speculation mechanisms.

Therefore, developing a method that can deeply understand the online learning behavior of students, thereby generating more accurate and personalized teaching advice has become an important research direction in the field of digital education.

Disclosure of Invention

In view of the foregoing, an object of the present application is to provide a big data pushing method and system based on digital education.

According to a first aspect of the present application, there is provided a big data pushing method based on digital education, applied to a digital education system, the method comprising:

performing correlation analysis on a plurality of priori learning events of a target digital education user on priori online learning nodes of a preset online learning node and a plurality of first learning mode data tracked by the preset online learning node to generate event correlation analysis data;

if the event correlation analysis data is used for determining that a target prior learning event which is not related to first learning mode data exists in the prior learning events, performing feature speculation based on the target prior learning event, and determining speculation information corresponding to a second learning mode data speculation learning node which is related to the target prior learning event before the preset online learning node, wherein the speculation information comprises speculation learning behaviors and speculation learning characterization parameters corresponding to the second learning mode data at the speculation learning node;

determining a presumption credible value of the presumption information based on the presumption learning characterization parameters corresponding to the presumption learning nodes, the learning characterization parameters corresponding to the final learning behavior in the target priori learning event and the continuous parameters of the target presumption learning behavior in the target priori learning event according to the second learning mode data; the target speculative learning behavior is used for representing the speculative learning behavior after the target learning behavior triggered by the terminal in the target priori learning event;

And if the presumed credible value is not smaller than the set credible value, carrying out learning event correlation configuration on the presumed learning behavior corresponding to the presumed learning node and the target priori learning event of the second learning mode data, and pushing each correlation configuration data serving as the digital education big data of the target digital education user into a corresponding big data service system.

In a possible implementation manner of the first aspect, the determining the speculative trusted value of the speculative information based on the speculative learning characterization parameter corresponding to the speculative learning node, the learning characterization parameter corresponding to the final learning behavior in the target prior learning event, and the persistent parameter of the target speculative learning behavior in the target prior learning event according to the second learning mode data includes:

determining the deviation degree of the second learning mode data between the speculative learning characterization parameters corresponding to the speculative learning nodes and the learning characterization parameters corresponding to the final learning behaviors in the target prior learning event;

determining a presumption confidence value of the presumption information based on the continuous parameters of the target presumption learning behavior and the deviation degree;

The presumption credibility value of the presumption information and the continuous parameter of the target presumption learning behavior are in reverse association, and the presumption credibility value of the presumption information and the deviation degree are in reverse association.

In a possible implementation manner of the first aspect, the plurality of first learning mode data is collected by a plurality of online monitoring components;

if it is determined that a target prior learning event which is not related to the first learning mode data exists in the plurality of prior learning events based on the event correlation analysis data, performing feature estimation based on the target prior learning event, and determining estimation information corresponding to a second learning mode data estimation learning node which is related to the target prior learning event before the preset online learning node, including:

if the monitoring field intervals of the plurality of online monitoring components at the preset online learning node are provided with shared field parts, determining that a target prior learning event which is irrelevant to first learning mode data exists in the plurality of prior learning events based on the event correlation analysis data, performing feature speculation based on the target prior learning event, and determining speculation information corresponding to a second learning mode data speculation learning node which is related to the target prior learning event before the preset online learning node.

In a possible implementation manner of the first aspect, the plurality of first learning mode data is collected according to a plurality of online monitoring components; the event correlation analysis data comprises first correlation characteristic information;

performing correlation analysis on a plurality of priori learning events of a priori online learning node of a target digital education user at the preset online learning node and a plurality of first learning mode data tracked by the preset online learning node, and generating event correlation analysis data, wherein the event correlation analysis data comprises:

determining a reference learning event from the plurality of prior learning events and determining reference learning pattern data from the plurality of first learning pattern data based on the monitoring service tags of the plurality of online monitoring components for each of the prior learning events and the monitoring service tags of the plurality of online monitoring components for each of the first learning pattern data;

determining a feature distance between the reference learning event and the reference learning mode data based on monitoring service tags of the plurality of online monitoring components for the reference learning event and monitoring service tags of the plurality of online monitoring components for the reference learning mode data;

And carrying out learning event correlation configuration on the reference learning event and the reference learning mode data based on the feature distance, and generating the first correlation feature information.

In a possible implementation manner of the first aspect, the determining the reference learning event from the plurality of prior learning events and the reference learning pattern data from the plurality of first learning pattern data based on the monitoring service tags of the plurality of online monitoring components for each of the prior learning events and the monitoring service tags of the plurality of online monitoring components for each of the first learning pattern data includes:

generating a monitoring service label sequence corresponding to each prior learning event based on the monitoring service labels of the online monitoring components for each prior learning event;

if a plurality of monitoring service tag sequences are provided with a target monitoring service tag sequence comprising monitoring service tags of the online monitoring components aiming at the first learning mode data, acquiring a priori learning event corresponding to the target monitoring service tag sequence as a reference learning event, and acquiring the first learning mode data as reference learning mode data.

In a possible implementation manner of the first aspect, the determining, based on the monitoring service tags of the plurality of online monitoring components for the reference learning event and the monitoring service tags of the plurality of online monitoring components for the reference learning mode data, a feature distance between the reference learning event and the reference learning mode data includes:

determining weight values of monitoring service tags of the plurality of online monitoring components aiming at the reference learning mode data in a monitoring service tag sequence corresponding to the reference learning event, and taking the weight values as service tag weight values corresponding to the reference learning mode data under the reference learning event;

and determining the ratio of the service tag weight value corresponding to the reference learning mode data under the reference learning event and the matching degree between the reference learning event and the reference learning mode data, and generating the characteristic distance between the reference learning event and the reference learning mode data.

In a possible implementation manner of the first aspect, the event correlation analysis data further includes second correlation characteristic information;

performing correlation analysis on a plurality of priori learning events of a priori online learning node of a target digital education user at a preset online learning node and a plurality of first learning mode data tracked by the preset online learning node to generate event correlation analysis data, and further comprising:

And if the first learning event which is irrelevant to the first learning mode data is determined in the plurality of prior learning events and the third learning mode data which is irrelevant to the prior learning event is determined in the plurality of first learning mode data based on the first relevance characteristic information, carrying out learning event relevance configuration on the first learning event and the third learning mode data based on the matching degree between the first learning event and the third learning mode data, and generating second relevance characteristic information.

In a possible implementation manner of the first aspect, the event correlation analysis data further includes third correlation characteristic information;

if the first correlation characteristic information and the second correlation characteristic information are based, determining that a second learning event which is not related to first learning mode data exists in the plurality of prior learning events and a fourth learning mode data which is not related to the prior learning event exists in the plurality of first learning mode data, relating the fourth learning mode data to a second learning event which is most matched with the fourth learning mode data for each fourth learning mode data, and generating third correlation characteristic information;

And discarding fifth learning mode data if it is determined that there is fifth learning mode data which is not related to the prior learning event in the plurality of second learning mode data based on the first correlation characteristic information, the second correlation characteristic information and the third correlation characteristic information.

In a possible implementation manner of the first aspect, the plurality of first learning mode data are collected on an online course education page;

if it is determined that the plurality of prior learning events have a second learning event that is not related to the first learning mode data and the plurality of first learning mode data have fourth learning mode data that is not related to the prior learning event based on the first correlation feature information and the second correlation feature information, for each fourth learning mode data, linking the fourth learning mode data to a second learning event that is most matched with the fourth learning mode data, generating third correlation feature information includes:

and if the online course education page is positioned in the learning mode sharing process, determining that a second learning event which is irrelevant to the first learning mode data exists in the multiple prior learning events and a fourth learning mode data which is irrelevant to the prior learning event exists in the multiple first learning mode data based on the first correlation characteristic information and the second correlation characteristic information, and relating the fourth learning mode data to the second learning event which is most matched with the fourth learning mode data aiming at each fourth learning mode data to generate third correlation characteristic information.

According to a second aspect of the present application, there is provided a digital education system comprising a machine-readable storage medium storing machine-executable instructions and a processor which, when executing the machine-executable instructions, implements the aforementioned digital education-based big data pushing method.

According to a third aspect of the present application, there is provided a computer-readable storage medium having stored therein computer-executable instructions that, when executed, implement the aforementioned big data push method based on digital education.

According to any one of the aspects, in the application, firstly, correlation analysis is performed on a plurality of priori learning events of a target digital education user at a preset online learning node and a plurality of first learning mode data tracked at the preset online learning node, so as to generate event correlation analysis data. If it is determined that there are target prior learning events that are not related to the first learning mode data based on the event correlation analysis data, the system further performs feature estimation based on the target prior learning events, and determines estimation information corresponding to the second learning mode data estimation learning node associated before the preset online learning node. The presumption information includes presumption learning behavior and presumption learning characterization parameters corresponding to the second learning mode data at the presumption learning node. Next, a speculative trusted value of the speculative information is determined based on the second learning mode data at the speculative learning characterization parameters corresponding to the speculative learning nodes, the learning characterization parameters corresponding to the final learning behavior in the target prior learning event, and the persistent parameters of the target speculative learning behavior in the target prior learning event. If the estimated trusted value is not less than the set trusted value, carrying out correlation configuration of the learning event, correlating the estimated learning behavior corresponding to the second learning mode data at the estimated learning node with the target priori learning event, and pushing each correlation configuration data serving as the digital education big data of the target digital education user to the corresponding big data service system. Therefore, more accurate personalized teaching advice is generated, and the online education effect is optimized.

That is, according to the embodiment of the application, the online learning behavior of the student is understood in a deeper and comprehensive manner by carrying out correlation analysis on the prior online learning event of the target digital education user at the preset online learning node and the first learning mode data tracked by the preset online learning node, generating event correlation analysis data and further carrying out feature speculation and correlation configuration, so that the learning behavior and learning mode of the student can be accurately identified and speculated, and high-quality teaching suggestions are generated. In addition, the priori learning events and learning modes of students are considered, so that the generated teaching advice can be personalized according to the specific requirements of each student, and further, the correlation configuration data is subjected to big data pushing, thereby being beneficial to optimizing the digital education pushing strategy. The learning behavior and learning mode of the students can be dynamically analyzed, and the prediction and configuration can be carried out according to the trusted value, so that the students can dynamically adjust along with the learning process of the students, and the most suitable teaching suggestion can be provided.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting in scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a big data pushing method based on digital education according to an embodiment of the present application;

fig. 2 is a schematic diagram of a component structure of a digital education system for implementing the above-mentioned big data pushing method based on digital education according to an embodiment of the present application.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below according to the drawings in the embodiments of the present application, and it should be understood that the drawings in the present application are only for the purpose of illustration and description, and are not intended to limit the protection scope of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this application, illustrates operations implemented in accordance with some embodiments of the present application. It should be understood that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Furthermore, one skilled in the art, under the direction of this application, may add at least one other operation to the flowchart, or may destroy at least one operation from the flowchart.

In addition, the described embodiments are only some, but not all, of the embodiments of the present application. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, correspond to the scope of protection of the present application, according to the embodiments of the present application.

Fig. 1 is a schematic flow chart of a big data pushing method and a big data pushing system based on digital education according to an embodiment of the present application, and it should be understood that, in other embodiments, the order of part of the steps in the big data pushing method based on digital education according to the present embodiment may be shared with each other according to actual needs, or part of the steps may be omitted or maintained. The big data pushing method based on digital education comprises the following detailed steps:

step S110, carrying out correlation analysis on a plurality of priori learning events of a target digital education user on a priori online learning node of a preset online learning node and a plurality of first learning mode data tracked on the preset online learning node to generate event correlation analysis data.

For example, assuming an online physical course, the preset online learning node includes topics such as mechanics, electromagnetism, and quantum physics. Student a has completed learning of mechanics and electromagnetism and entered into a learning node of quantum physics. At this time, the behavior patterns of the student a in the previous learning nodes (such as mechanics and electromagnetics) may be recorded, such as time they spent, number of exercises completed, error type, etc., and correlation analysis may be performed with the first learning pattern data of the quantum physical learning node (possibly from the learning behaviors of other users at the quantum physical node), to generate event correlation analysis data.

That is, in the example of the above-described physical course, the target digital education user refers to student a who wants to provide personalized teaching advice by analyzing his learning behavior and pattern.

The preset online learning node refers to various theme parts of online physical courses, such as mechanics, electromagnetism, quantum physics and the like. Each topic section can be considered a learning node.

The multiple prior learning events of the prior online learning node refer to the learning behavior of student a on both the online learning node before quantum physics is completed, i.e., mechanics and electromagnetism. These learning activities may include the time he spends solving a problem, the speed at which the job is completed, the accuracy of answering the problem, etc.

The first learning mode data of the preset online learning node tracking refers to learning behavior data of the quantum physical node collected from other students, such as average time of solving the problem, error type and the like.

And carrying out correlation analysis on learning events of the student A at mechanical and electromagnetic nodes and learning mode data tracked at quantum physical nodes to generate event correlation analysis data, wherein the event correlation analysis data means that the learning behaviors of the student A at the mechanical and electromagnetic nodes can be compared with the learning behaviors of other students at the quantum physical nodes to check whether correlation exists. If student A's learning behavior at the mechanical and electromagnetic nodes is significantly different from the successful learning behavior of other students at the quantum-physical nodes, it is believed that student A may experience difficulty at the quantum-physical nodes and follow-up speculation and advice accordingly.

Step S120, if it is determined that the plurality of prior learning events have a target prior learning event that is not related to the first learning mode data based on the event correlation analysis data, performing feature estimation based on the target prior learning event, and determining estimation information corresponding to a second learning mode data estimation learning node associated with the target prior learning event before the preset online learning node, where the estimation information includes an estimation learning behavior and an estimation learning characterization parameter corresponding to the second learning mode data at the estimation learning node.

For example, if the analysis finds that certain learning behaviors of student a at the mechanical nodes (target a priori learning events) have no correlation with the learning patterns of the quantum physical nodes, then feature projections can be made on these target a priori learning events. For example, if student A is found to be overly focused on the application of formulas at the mechanical nodes and neglect the understanding of theoretical knowledge, while a deeper theoretical understanding is needed at the quantum physical nodes, then it can be speculated that student A may experience difficulties at the quantum physical nodes.

That is, in this example of physical lessons, the event correlation analysis data refers to data obtained by comparing learning behavior of student a on mechanical and electromagnetic nodes with learning behavior of other students on quantum physical nodes.

If certain learning behaviors of student a on mechanical and electromagnetic nodes (e.g., overreliance on formula application and neglecting theoretical understanding) are different from other learning patterns of students who successfully complete quantum physical learning, then these prior learning events are considered to be target prior learning events that are not related to the first learning pattern data.

At this time, feature estimation may be performed based on these target prior learning events. That is, it may be attempted to predict what learning behavior would be exhibited in the quantum physical node if student a continued his learning behavior at the mechanical and electromagnetic nodes (e.g., over-dependent formula application).

The presumed information includes predicted learning behavior of student a at the quantum physical node (e.g., may encounter difficulty in understanding theoretical knowledge) and characterization parameters of this prediction (e.g., degree of difficulty, progress of learning that may be affected, etc.). Such speculation information is used for subsequent determination of speculative trusted values and learning event correlation configurations.

Step S130, determining a speculative trusted value of the speculative information based on the speculative learning characterization parameter corresponding to the speculative learning node, the learning characterization parameter corresponding to the final learning behavior in the target prior learning event, and the persistent parameter of the target speculative learning behavior in the target prior learning event. The target speculative learning behavior is used to characterize the speculative learning behavior after the target learning behavior triggered by the end in the target prior learning event.

For example, the confidence level of the speculation may be determined based on the last learning behavior of student a at the mechanical node, the inferred learning behavior of the quantum-physical node, and the time for which the speculation may last. For example, if student a is too much focused on the behavior of the formula application on the mechanical nodes until it is last, then he encounters difficult speculation on the quantum-physical nodes is more trusted.

That is, in this example of physical course, the speculative learning characterization parameter corresponding to the second learning mode data at the speculative learning node refers to the predicted learning behavior of student a that may occur at the quantum physical node and the characteristics of the learning behavior. For example, it may be predicted that student a encounters difficulty in understanding quantum physics theory due to over-reliance on formula application.

The learning characterization parameters corresponding to the final learning behavior in the target prior learning event refer to the learning behavior of the student a at the end of the previous learning node (electromagnetics) and the characteristics of this learning behavior. For example, if student A still relies too much on formula application at the end of the electromagnetism section, then this learning behavior is the end learning behavior. The persistence parameters of the target-speculative learning behavior may then include the time, frequency, etc. that student a continues to rely excessively on the application of the formula.

The confidence value of the speculation information, that is, the confidence that the prediction student a may encounter difficulty on the quantum physical node, may be determined based on the above three parameters. If all three parameters show that student A is likely to experience difficulty in understanding quantum physics theory, then the presumed confidence value will be high.

The target speculative learning behavior refers to the possible learning behavior (difficulty in understanding theoretical knowledge) of the student at the quantum physical node, which is predicted based on the final learning behavior (overdependent formula application) of the student a at the electromagnetic node. This target speculates that learning behavior will be used for subsequent learning event correlation configurations.

And step S140, if the presumed credible value is not less than the set credible value, carrying out learning event correlation configuration on the presumed learning behavior corresponding to the presumed learning node and the target priori learning event of the second learning mode data, and pushing each correlation configuration data serving as the digital education big data of the target digital education user into a corresponding big data service system.

For example, if the presumed confidence value is above a certain threshold, the system correlates student A's learning behavior at the mechanical node with the presumed learning behavior of the quantum physical node and sends this information as student A's digital educational big data to the big data service system. For example, if the confidence of the speculation is very high, the system may alert student a to enhance learning of theoretical knowledge before starting quantum physical nodes, or recommend additional resources to him that deepen theoretical understanding.

That is, in the example of physical lessons, the speculative trusted value refers to the confidence of predictions that student a may experience difficulty at the quantum physical node. The credibility is calculated based on parameters such as learning behavior of the student A at the end of the electromagnetic node, predicted learning behavior of the quantum physical node, and possible duration of the behavior. If this speculation confidence value is not less than the set threshold (e.g., 0.7 or 70%), then this means that the current prediction result is relatively accurate. At this time, the next operation may be performed, that is, the learning behavior of the student a at the electromagnetic node (e.g., the application of the over-dependent formulas) is associated with the predicted learning behavior of the quantum physical node (e.g., the difficulty in understanding the theoretical knowledge may be encountered). That is, such information is recorded: student a may experience difficulty in understanding theoretical knowledge while learning quantum physics due to excessive dependence on formula application. These associated data are then sent to the big data service system as the digital educational big data for student a. Where these digital educational data can be further analyzed to generate more personalized learning advice, such as prompting student a to enhance learning of theoretical knowledge prior to starting quantum physical nodes, or recommending him additional resources that deepen theoretical understanding. Thus, the student A can be better helped to learn effectively.

Based on the steps, firstly, carrying out correlation analysis on a plurality of priori learning events of a target digital education user at a preset online learning node and a plurality of first learning mode data tracked at the preset online learning node, and generating event correlation analysis data. If it is determined that there are target prior learning events that are not related to the first learning mode data based on the event correlation analysis data, the system further performs feature estimation based on the target prior learning events, and determines estimation information corresponding to the second learning mode data estimation learning node associated before the preset online learning node. The presumption information includes presumption learning behavior and presumption learning characterization parameters corresponding to the second learning mode data at the presumption learning node. Next, a speculative trusted value of the speculative information is determined based on the second learning mode data at the speculative learning characterization parameters corresponding to the speculative learning nodes, the learning characterization parameters corresponding to the final learning behavior in the target prior learning event, and the persistent parameters of the target speculative learning behavior in the target prior learning event. If the estimated trusted value is not less than the set trusted value, carrying out correlation configuration of the learning event, correlating the estimated learning behavior corresponding to the second learning mode data at the estimated learning node with the target priori learning event, and pushing each correlation configuration data serving as the digital education big data of the target digital education user to the corresponding big data service system. Therefore, more accurate personalized teaching advice is generated, and the online education effect is optimized.

In one possible implementation, step S130 may include:

step S131, determining a degree of deviation of the second learning mode data between the speculative learning characterization parameter corresponding to the speculative learning node and the learning characterization parameter corresponding to the final learning behavior in the target prior learning event.

For example, in the physical course example, a degree of deviation between a predicted learning behavior (e.g., student a may experience difficulty in understanding quantum physical theory) and his actual learning behavior at the end of the electromagnetic node (e.g., over-dependent formula application) may be determined. If the difference between the two is large, the deviation degree is high; conversely, if the two are close, the degree of deviation will be lower.

For example, euclidean distance may be used to measure the difference between the extrapolated learning characterization parameters (i.e., the predicted degree to which student A may experience difficulty at the quantum-physical node) and the learning characterization parameters corresponding to the final learning behavior (i.e., the degree to which student A is over-dependent on the application of the formula at the end of the electromagnetic node). Assuming that both are quantized to values of 0-10, where 10 represents the highest degree, the degree of deviation can be calculated as the euclidean distance between them, i.e., absolute value | predictive learning characterization parameter-learning characterization parameter corresponding to the end learning behavior.

Step S132, determining a presumption confidence value of the presumption information based on the continuation parameter of the target presumption learning behavior and the degree of deviation. Wherein the presumption credibility value of the presumption information and the continuous parameter of the target presumption learning behavior are in reverse association, and the presumption credibility value of the presumption information and the deviation degree are in reverse association.

For example, after determining the degree of deviation, it is also necessary to consider how long student A's behavior that is excessively dependent on the application of the formula may last. If he still over depends on the formula application at the end of the electromagnetic node, the persistence parameter of this behavior will be high. Combining these two factors, a trusted value for the prediction information can be determined.

If the duration of the behavior of student a excessively depends on the formula application is long (i.e., the duration parameter is high), or the difference between the predicted learning behavior and the actual learning behavior is large (i.e., the deviation degree is high), the reliability of predicting that student a may encounter difficulty at the quantum physical node is reduced. That is, the higher the persistence parameter and the degree of deviation, the lower the predicted confidence value. This is because if the learning behavior of student a is strong in duration, or his learning behavior is greatly different from the predicted learning behavior, then his possibility of changing the learning behavior is reduced, resulting in an increase in inaccuracy of the prediction.

For example, a speculatively trusted value may be defined as a number between 0 and 1, where 1 represents completely trusted and 0 represents completely untrusted. Assuming that the duration parameter of the target speculative learning behavior (i.e., the length of time that student a may be excessively dependent on the behavior of the formula application) is also quantified as a value of 0-10, then the speculative confidence value may be calculated as 1- (0.5 x bias/10+0.5 x duration parameter/10). This formula assumes that the influence of the bias and persistence parameters on the extrapolated trusted value is equal and that they are both negative effects, i.e. the larger the value the smaller the trusted value.

The above is just a simplified example, and the actual computation may involve more complex statistical methods, such as using multiple linear regression, logistic regression, or machine learning models to account for interactions and non-linear relationships of various factors. However, the goal of these calculations is the same in any case: the reliability of the predictions of future learning behavior is evaluated based on existing information.

For example, in some other examples, if more parameter factors are considered, it may be desirable to add a parameter reflecting the overall performance of the student at the previous node (e.g., average score M), and a parameter reflecting the student's enthusiasm or engagement (e.g., activity L on an online learning platform). Then the formula may be rewritten as:

assuming that P represents the "speculative learning characterization parameter" and a represents the "learning characterization parameter corresponding to the final learning behavior", the degree of deviation D can be calculated as:

D = |P - A| / (1 + M/100)

here it is assumed that the average score M is a value of 0-100 and that it has a negative influence on the degree of deviation: the higher the score, the smaller the degree of deviation.

Assuming that T represents a "sustained parameter of target-speculative learning behavior", then the speculative trusted value C can be calculated as:

C = 1 - 0.5 * (D/10 + T/10) + L/100

Here, it is assumed that the liveness L is a value of 0-100 and that it has a positive influence on the presumed trusted value: the higher the liveness, the greater the trusted value.

The above formula is an extended example, and in practical applications, more complex modeling may be required according to actual situations and data.

In one possible embodiment, the plurality of first learning mode data is collected by a plurality of online monitoring components.

In the physical course example, the online monitoring component may be various functions of the online educational platform, such as test performance of students, learning time, frequency of participation in discussions, and the like. These online monitoring components collect learning behavior data of student a on mechanical and electromagnetic nodes and learning behavior data of other students on quantum physical nodes.

It is assumed that the on-line monitoring component collects data such as test results, learning time, etc. of student a at mechanical and electromagnetic nodes, and collects the same data of other students at quantum physical nodes. That is, the monitoring field intervals of these online monitoring components (i.e., the types of data they collect) have shared field portions (i.e., test achievements, learning times, etc.). Then, the learning behavior of student a on the mechanical and electromagnetic nodes can be compared with the learning behavior of other students on the quantum physical nodes, and if a significant difference is found between the learning behavior of student a and the behavior of the student who successfully completes the quantum physical learning, these learning behaviors are considered as target prior learning events unrelated to the first learning mode data. Then, feature speculation is performed based on the prior learning events, and possible learning behaviors of the student A at the quantum physical node are predicted.

In one possible embodiment, the plurality of first learning mode data is collected in accordance with a plurality of on-line monitoring components. The event correlation analysis data includes first correlation characteristic information.

For example, in the physical course example, the online monitoring component may be an online educational platform capable of tracking and recording learning behavior data of student a on mechanical and electromagnetic nodes, as well as learning behavior data of other students on quantum physical nodes. Such data includes learning time, number of exercises completed, type of error, etc.

Step S110 may include:

step S111, determining a reference learning event from the multiple prior learning events and determining reference learning pattern data from the multiple first learning pattern data based on the monitoring service tags of the multiple online monitoring components for each of the prior learning events and the monitoring service tags of the multiple online monitoring components for each of the first learning pattern data.

For example, the reference learning event and the reference learning mode data may be determined based on data collected by the online monitoring component. For example, the reference learning event may be the behavior of student a over-relying on the application of a formula on an electromagnetic node, while the reference learning pattern data may be the learning behavior of other students on a quantum physical node.

Step S112, determining a feature distance between the reference learning event and the reference learning mode data based on the monitoring service tags of the plurality of online monitoring components for the reference learning event and the monitoring service tags of the plurality of online monitoring components for the reference learning mode data.

For example, feature distances between reference learning events and reference learning pattern data may be calculated, which may be calculated based on their differences in different features. For example, if student a relies heavily on formula applications at the electromagnetic node, while other students focus more on theoretical understanding at the quantum physical node, the feature distance between the two may be larger.

Step S113, performing learning event correlation configuration on the reference learning event and the reference learning mode data based on the feature distance, and generating the first correlation feature information.

For example, the correlation between the reference learning event and the reference learning mode data may be analyzed based on the calculated feature distance, and the correlation feature information may be generated. For example, if the feature distance is large, which indicates that there is a significant difference between the learning behavior of student a at the electromagnetic node and the learning behavior of other students at the quantum physical node, the correlation feature information may show a low correlation between the two.

In one possible implementation, step S111 may include:

step S1111, based on the monitoring service labels of the multiple online monitoring components for each prior learning event, a monitoring service label sequence corresponding to each prior learning event is generated.

For example, a sequence of monitoring service tags for each a priori learning event may be generated from data collected by the online monitoring component. For example, for each learning behavior of student a (e.g., number of exercises performed, type of errors, etc.) on the mechanics and electromagnetics nodes, a corresponding monitoring service tag is generated and organized into a sequence in a certain order.

In step S1112, if a plurality of monitoring service tag sequences have a target monitoring service tag sequence including monitoring service tags of the plurality of online monitoring components for the first learning mode data, a priori learning event corresponding to the target monitoring service tag sequence is obtained as a reference learning event, and the first learning mode data is obtained as reference learning mode data.

For example, a monitoring service tag sequence may be found that includes monitoring service tags for the first learning mode data (i.e., learning behavior data of other students on the quantum physical node) by the online monitoring component. For example, if there is a tag sequence representing the behavior of student a over-dependent on the application of the formula on the electromagnetic node, and this behavior is also present in the learning behavior of other students on the quantum-physical node, then this tag sequence is selected as the target monitoring service tag sequence, and the corresponding learning event (over-dependent on the application of the formula) and learning pattern data (learning behavior of other students on the quantum-physical node) are selected as the reference learning event and the reference learning pattern data.

In one possible implementation, step S112 may include:

step S1121, determining a weight value of the monitoring service tag of the plurality of online monitoring components for the reference learning mode data in the monitoring service tag sequence corresponding to the reference learning event, as a service tag weight value corresponding to the reference learning mode data under the reference learning event.

For example, service tag weight values of monitoring service tags of the online monitoring component for reference learning pattern data (i.e., learning behavior of other students on quantum physical nodes) in a sequence of monitoring service tags corresponding to a reference learning event (such as behavior of student a over-dependent on formula application on electromagnetic nodes) may be calculated. This service tag weight value reflects the importance of the reference learning mode data in the reference learning event.

Step S1122, determining a ratio of the service tag weight value corresponding to the reference learning mode data under the reference learning event to the matching degree between the reference learning event and the reference learning mode data, and generating a feature distance between the reference learning event and the reference learning mode data.

For example, a degree of matching between the reference learning event and the reference learning pattern data may be calculated, and then a service tag weight value may be calculated by a ratio operation with this degree of matching to obtain the feature distance. This feature distance reflects the similarity or difference between the reference learning event and the reference learning mode data. For example, if the feature distance is large, it may be stated that there is a significant difference between student a's learning behavior on the electromagnetic node and other students' learning behavior on the quantum-physical node.

In one possible embodiment, the event correlation analysis data further includes second correlation characteristic information.

Step S110 may further include: and if the first learning event which is irrelevant to the first learning mode data is determined in the plurality of prior learning events and the third learning mode data which is irrelevant to the prior learning event is determined in the plurality of first learning mode data based on the first relevance characteristic information, carrying out learning event relevance configuration on the first learning event and the third learning mode data based on the matching degree between the first learning event and the third learning mode data, and generating second relevance characteristic information.

For example, those learning events that are not related to the first learning mode data (i.e., the learning behavior of other students on the quantum physical node) and those learning mode data that are not related to the prior learning events (i.e., the learning behavior of student a on the mechanical and electromagnetic nodes) may be found based on the first correlation characteristic information that has been calculated. For example, if student a spends a lot of time reading theory at the mechanics node, but is overly dependent on formula application at the electromagnetics node, then both behaviors may be considered unrelated learning events. Then, the matching degree between these uncorrelated learning events and the learning pattern data is calculated, and correlation configuration is performed based on this matching degree, thereby generating second correlation characteristic information. For example, if student a reads the theoretical time on the mechanical node and other students read the theoretical time on the quantum-physical node with a high degree of matching, the correlation configuration between the two may result in a higher correlation characteristic value as the second correlation characteristic information.

In one possible embodiment, the event correlation analysis data further includes third correlation characteristic information.

Step S110 may further include: if it is determined that a second learning event which is not related to the first learning mode data exists in the plurality of prior learning events and a fourth learning mode data which is not related to the prior learning event exists in the plurality of first learning mode data based on the first correlation characteristic information and the second correlation characteristic information, the fourth learning mode data is related to a second learning event which is most matched with the fourth learning mode data for each fourth learning mode data, and third correlation characteristic information is generated. And discarding fifth learning mode data if it is determined that there is fifth learning mode data which is not related to the prior learning event in the plurality of second learning mode data based on the first correlation characteristic information, the second correlation characteristic information and the third correlation characteristic information.

For example, those learning events and learning pattern data that are not related to the first learning pattern data (i.e., the learning behavior of other students on quantum physical nodes) and the prior learning events (i.e., the learning behavior of student a on mechanical and electromagnetic nodes) can be found based on the first and second correlation characteristic information that has been calculated. Then, for each of the found uncorrelated fourth learning pattern data, the second learning event that best matches it is found and correlated, thereby generating third correlation characteristic information. Then, according to all the correlation characteristic information which has been calculated, fifth learning mode data which are not related to the prior learning event are found out, and the data are discarded. This means that these discarded data are no longer considered in the subsequent analysis and advice generation process.

In one possible embodiment, the plurality of first learning mode data is collected on an online class education page.

In the above embodiment, if the online course education page is located in the learning mode sharing process, and based on the first correlation characteristic information and the second correlation characteristic information, it is determined that the plurality of prior learning events have a second learning event that is not related to the first learning mode data and the plurality of first learning mode data have fourth learning mode data that is not related to the prior learning event, for each fourth learning mode data, the fourth learning mode data is linked to a second learning event that is most matched with the fourth learning mode data, and third correlation characteristic information is generated.

For example, first confirm that the first learning mode data collected (i.e., the learning behavior of other students on the quantum physical node) is obtained through an online course education page. For example, the data may come from an online educational platform that provides physical lessons and records the student's behavior during the learning process. If the online curriculum educational page is located within a shared learning model process, the system locates and correlates learning events and learning model data that are not related to the first learning model data (i.e., learning behavior of other students on quantum physical nodes) and prior learning events (i.e., learning behavior of student A on mechanical and electromagnetic nodes) based on the first and second correlation characteristic information that has been computed, thereby generating third correlation characteristic information.

This shared learning mode process can be understood as a background service of an online educational platform that is responsible for managing and analyzing learning mode data for all online courses. In this way, the system can conduct learning behavior analysis in a wider range, thereby providing more comprehensive and more accurate personalized teaching advice.

Fig. 2 schematically illustrates a digital educational system 100 that can be used to implement various embodiments described in the present application.

For one embodiment, fig. 2 shows a digital educational system 100, the digital educational system 100 having at least one processor 102, a control module (chipset) 104 coupled to at least one of the (at least) processors 102, a memory 106 coupled to the control module 104, a non-volatile memory (NVY)/storage device 108 coupled to the control module 104, at least one input/output device 110 coupled to the control module 104, and a network interface 112 coupled to the control module 104.

The processor 102 may include at least one single-core or multi-core processor, and the processor 102 may include any combination of general-purpose processors or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In an alternative embodiment, the digital educational system 100 can function as a server device such as a gateway as described in the examples herein.

For one embodiment, fig. 2 shows a digital educational system 100, the digital educational system 100 having at least one processor 102, a control module (chipset) 104 coupled to at least one of the (at least one) processors 102, a memory 106 coupled to the control module 104, a non-volatile memory (NVM)/storage 108 coupled to the control module 104, at least one input/output device 110 coupled to the control module 104, and a network interface 112 coupled to the control module 104.

In an alternative embodiment, the digital educational system 100 can include at least one computer-readable medium (e.g., memory 106 or NVM/storage 108) having instructions 114 and at least one processor 102, in combination with the at least one computer-readable medium, configured to execute the instructions 114 to implement the modules to perform the actions described in the present disclosure.

For one embodiment, the control module 104 may include any suitable interface controller to provide any suitable interface to at least one of the (at least one) processor 102 and/or any suitable device or component in communication with the control module 104.

The control module 104 may include a memory controller module to provide an interface to the memory 106. The memory controller modules may be hardware modules, software modules, and/or firmware modules.

The memory 106 may be used, for example, to load and store data and/or instructions 114 for the digital educational system 100. For one embodiment, memory 106 may comprise any suitable volatile memory, such as, for example, a suitable DRAM. In an alternative embodiment, memory 106 may comprise a double data rate type four synchronous dynamic random access memory (DDR 4 SDRAM).

For one embodiment, the control module 104 may include at least one input/output controller to provide an interface to the NVM/storage 108 and the (at least one) input/output device 110.

For example, NVM/storage 108 may be used to store data and/or instructions 114. NVM/storage 108 may include any suitable nonvolatile memory (e.g., flash memory) and/or may include any suitable (at least one) nonvolatile storage (e.g., at least one Hard Disk Drive (HDD), at least one Compact Disc (CD) drive, and/or at least one Digital Versatile Disc (DVD) drive).

The NVM/storage 108 may include a storage resource that is physically part of the device on which the digital educational system 100 is installed, or it may be accessible by the device, or it may not be necessary as part of the device. For example, NVM/storage 108 may be accessed via (at least one) input/output device 110 over a network.

The (at least one) input/output device 110 may provide an interface for the digital educational system 100 to communicate with any other suitable device, the input/output device 110 may include a communication component, pinyin component, online monitoring component, etc. The network interface 112 may provide an interface for the digital educational system 100 to communicate in accordance with at least one network, and the digital educational system 100 may communicate wirelessly with at least one component of a wireless network in accordance with any of at least one wireless network standard and/or protocol, such as accessing a wireless network in accordance with a communication standard, or a combination thereof.

For one embodiment, at least one of the (at least one) processor 102 may be loaded with logic of at least one controller (e.g., memory controller module) of the control module 104. For one embodiment, at least one of the (at least one) processor 102 may be loaded together with logic of at least one controller of the control module 104 to form a system level load. For one embodiment, at least one of the (at least one) processor 102 may be integrated on the same die as the logic of at least one controller of the control module 104. For one embodiment, at least one of the (at least one) processor 102 may be integrated on the same die with logic of at least one controller of the control module 104 to form a system on chip (SoC).

In various embodiments, the digital educational system 100 may be, but is not limited to: a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet, a netbook, etc.), among other terminal devices. In various embodiments, the digital educational system 100 may have more or fewer components and/or different architectures. For example, in one alternative embodiment, the digital educational system 100 includes at least one camera, a keyboard, a Liquid Crystal Display (LCD) screen (including a touch screen display), a non-volatile memory port, multiple antennas, a graphics chip, an Application Specific Integrated Circuit (ASIC), and a speaker.

The foregoing has outlined rather broadly the more detailed description of embodiments of the present application, wherein specific examples are provided herein to illustrate the principles and embodiments of the present application, the above examples being provided solely to assist in the understanding of the methods of the present application and the core ideas thereof; meanwhile, as those skilled in the art will have modifications in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims

1. A big data push method based on digital education, characterized in that it is applied to a digital education system, the method comprising:

Performing correlation analysis on a plurality of priori learning events of a target digital education user at a priori online learning node of a preset online learning node and a plurality of first learning mode data tracked at the preset online learning node to generate event correlation analysis data, wherein the first learning mode data refer to learning behavior data collected from other digital education users at the preset online learning node;

If the presumed credible value is not smaller than the set credible value, carrying out learning event correlation configuration on presumed learning behaviors corresponding to the presumed learning nodes and the target priori learning events of the second learning mode data, and pushing each correlation configuration data serving as digital education big data of the target digital education user into a corresponding big data service system;

the determining, based on the second learning mode data, the speculative learning characterization parameter corresponding to the speculative learning node, the learning characterization parameter corresponding to the final learning behavior in the target prior learning event, and the persistent parameter of the target speculative learning behavior in the target prior learning event, the speculative trusted value of the speculative information includes:

2. The digital education-based big data pushing method according to claim 1, wherein the plurality of first learning mode data are collected by a plurality of on-line monitoring components;

3. The digital education-based big data pushing method according to claim 1, wherein the plurality of first learning mode data are collected according to a plurality of on-line monitoring components; the event correlation analysis data comprises first correlation characteristic information;

4. The digital education-based big data pushing method according to claim 3, wherein the determining the reference learning pattern data from the plurality of prior learning events and the reference learning pattern data from the plurality of first learning pattern data based on the monitoring service tags of the plurality of online monitoring components for each of the prior learning events and the monitoring service tags of the plurality of online monitoring components for each of the first learning pattern data comprises:

5. The digital education-based big data pushing method according to claim 3, wherein the determining the feature distance between the reference learning event and the reference learning mode data based on the monitoring service tags of the plurality of on-line monitoring components for the reference learning event and the monitoring service tags of the plurality of on-line monitoring components for the reference learning mode data comprises:

6. The digital education-based big data pushing method according to claim 3, wherein the event correlation analysis data further includes second correlation characteristic information;

7. The digital education-based big data pushing method according to claim 6, wherein the event correlation analysis data further includes third correlation characteristic information;

8. The digital education-based big data pushing method according to claim 7, wherein the plurality of first learning mode data are collected on an online course education page;

9. A digital education system comprising a processor and a computer readable storage medium storing machine executable instructions that when executed by the processor implement the digital education based big data pushing method of any one of claims 1-8.