Optimization of teaching methods and allocation of learning resources under the background of big data

Abstract

In the context of the wide application of big data technology, it is particularly important to optimize the allocation of teaching methods and learning resources. This study first expounds the key role of big data in the optimization of teaching methods and the allocation of learning resources, and emphasizes how big data technology promotes the transformation and development of education and teaching models. Based on the analysis of traditional models of teaching method optimization and learning resource allocation, this study proposes a new model driven by big data. By accurately identifying students’ learning needs and behavior patterns, the model optimizes teaching methods and allocation of learning resources. This study introduces the whole process of data collection, cleaning, analysis and modeling. In the process, it shows how big data can be integrated, analyzed, and applied to further support the construction and validation of models. Through empirical research and effect evaluation, this study proves the validity of the model of teaching method optimization and learning resource allocation driven by big data, and demonstrates how big data can promote educational equity and improve educational quality.

Keywords

Big data teaching method optimization learning resource allocation data-driven model

1. Introduction

In the context of today’s information society, big data has penetrated into various industries and fields, including education. In fact, big data is not only a technical topic, but also a tool and way of thinking that can optimize teaching methods and the allocation of learning resources. For the field of education research, how to optimize teaching methods and resource allocation with the power of big data has become an issue that cannot be ignored. Because of many factors, such as lack of individuation and intelligence, traditional teaching methods and resource allocation often can not meet the diversified needs of modern education.

Big data can be defined here as the rapid acquisition, storage, and analysis of large amounts of data from a variety of sources and in a variety of formats. This data can be structured or unstructured, but the point is that by analyzing this data, useful information and insights can be gained. Big data technology includes data collection, data cleaning, data storage, data analysis and other links. In the field of education, big data can be used to personalize instruction, optimize classroom interaction, improve teaching quality, and more accurately allocate educational resources. In terms of theoretical basis, this study will rely on data-driven teaching model and resource allocation strategy, combined with teaching theories such as constructivism and behaviorism, to build and verify the corresponding formula model. This model not only helps to understand how big data is changing teaching methods and resource allocation, but also illustrates complex big data concepts in a way that is easier to understand. In order to support the validity of the model, we will conduct a series of data pre-processing, including data cleaning and data analysis, and select appropriate samples for empirical research.

In the field of educational research, teaching methods and allocation of learning resources are two closely related aspects that together form the basis of a quality educational experience. Teaching methods are not only concerned with teaching strategies and techniques inside and outside the classroom, but also how learning resources are effectively allocated and utilized. For example, adopting a personalized teaching approach may require a specific configuration of learning resources to meet the individual needs of different students, such as advanced textbooks, online educational software, or personalized assignments. Conversely, if resources are limited or unevenly distributed, even the most advanced teaching methods may not reach their maximum utility.

In the context of big data, the relationship between the two becomes clearer. Big data can not only optimize existing teaching methods, but also enable more precise allocation of educational resources through real-time analysis of student behavior and feedback. This means that educators can use the information gathered from large amounts of data to adjust teaching strategies and resource allocation plans to better meet the actual needs of students. This is especially important in distance or blended education models, which often involve more complex and decentralized resource allocation issues.

In short, the optimization of teaching methods and allocation of learning resources in the context of big data is an interdisciplinary and highly integrated research field. Big data not only provides a new perspective on teaching and learning, but also provides us with rich data resources and efficient data analysis tools to support more personalized and intelligent education. Research like this aims to fill the gap between traditional methods and modern needs, bringing smarter and more effective solutions to the education community. By delving into the specific application of big data in teaching and resource allocation, research can not only gain a more comprehensive understanding of big data concepts and technologies, but also more accurately grasp the challenges and opportunities facing the education sector.

In terms of big data research on education and resource allocation, several studies have explored different aspects of teaching optimization and resource allocation in depth. Sreethar et al. proposed a prioritization based resource allocation algorithm, which is mainly applied in wireless networks, but its core concept can be used for reference in educational resource allocation, especially in online or mixed teaching environments [1]. In addition, Hou et al. used the adaptive teaching-learning optimization method to study the teaching evaluation of WebGIS courses, showing that big data and optimization algorithms can be used to evaluate and improve teaching effects [2]. For research on optimization of teaching methods, Quadri et al. applied a hybrid teaching-learning optimization technology in power distribution systems, showing that optimization algorithms can work under multi-objective and multi-constraint conditions, which provides useful insights into complex curriculum scheduling and resource allocation problems in educational scenarios [3]. At the same time, Li et al. studied teaching-learning optimization methods with multi-factor characteristics, especially diversity and triangular cooperation mechanism, which provided a new perspective and method for realizing personalized and intelligent teaching [4]. In terms of big data and machine learning, Gupta and Pawar proposed a structurally efficient multi-mode deep convolutional neural network, which added adaptive group teaching optimization. This research can not only be applied to complex tasks such as image recognition, but also provide possibilities for the design of personalized learning paths in the field of education [5]. In the research related to the allocation of educational resources in the aspect of network and cloud computing, Saidi and Bardou discussed the challenges and opportunities of task scheduling and virtual machine placement to resource allocation in cloud computing, which provided theoretical support for the dynamic and elastic allocation of educational resources [6]. To sum up, the existing literature provides a wealth of theoretical and empirical studies on the optimization of teaching methods and allocation of learning resources, but most of them focus on specific application scenarios or technical fields. Therefore, synthesizing and expanding these studies in the context of big data will provide a more comprehensive and in-depth understanding of educational research.

The main purpose of this study is to optimize teaching methods and learning resource allocation using big data technology. By studying how big data affects and changes teaching methods and resource allocation, corresponding mathematical models are established and quantitative analysis is carried out, so as to explore a new model of data-driven teaching and resource allocation. For teaching methods, the research will analyze the limitations of its traditional model and explore how it can be optimized using big data technology. As for the allocation of learning resources, the research will start from the traditional allocation mode and study how to use big data for more effective resource allocation. At the same time, the practicability and validity of the established model will be verified by empirical analysis.

The significance of this study is mainly reflected in the optimization of teaching methods, which will help to improve the teaching effect, better meet the learning needs of students, and promote the innovation of teaching methods. For the allocation of learning resources, by optimizing the allocation of resources, we can make more effective use of educational resources and improve the efficiency of education [7, 8, 9]. This study will also provide data support for educational decision-makers and provide basis for educational management and teaching reform. The results of this study are expected to provide reference and inspiration for the application of big data in other fields, so as to promote the application of big data technology in a wider range of fields.

2. Application of big data in teaching method optimization

2.1 Traditional model of teaching methods

The traditional teaching methods are usually teacher-centered, emphasizing the links between teacher teaching and student learning, and lack of adaptability to students’ individual differences. It usually contains the following core elements, as shown in the Fig. 1.

Figure 1.

Core elements of traditional teaching methods.

A In this model, teaching evaluation is usually based on standardized tests and examinations, which not only cannot fully reflect students’ learning progress, but also cannot give timely feedback for teaching adjustment. Such limitations become particularly evident in the trend of big data and personalized learning. In the traditional mode, due to the lack of effective data analysis tools, it is difficult for teachers to immediately know the learning status of each student, so as to make fine teaching adjustments. Therefore, it has become an important task of current educational research to explore new teaching method models that adapt to the era of big data.

2.2 The necessity of teaching method optimization

In the modern educational environment, the necessity of optimization of teaching methods is obvious [10]. The main reasons can be explained from the following aspects:

(1)
Personalized teaching needs: Each student has his or her own unique learning style and ability, and traditional teaching methods are difficult to meet the needs of all students. Optimizing teaching methods and realizing personalized teaching can better meet students’ learning needs and improve learning results.
(2)
Optimization of teaching resources: Optimization of teaching methods can not only improve teaching efficiency, but also realize optimal utilization of teaching resources. For example, the use of big data technology can accurately understand each student’s learning progress and problems, timely adjustment of teaching strategies, so that teaching resources can be more efficient use [11, 12].
(3)
Educational equity: Optimizing teaching methods can promote educational equity. Using big data technology, students can be evaluated more fairly and fairly, avoiding the bias and injustice existing in traditional evaluation methods.
(4)
Improving the quality of education: Optimizing teaching methods can improve the quality of education. For example, through data analysis, teachers can learn which teaching methods work better for students and adjust their teaching strategies.
(5)
Coping with social changes: With the rapid development of society, the demand for talents is also changing [13]. The optimization of teaching methods can make education better adapt to social changes and cultivate talents more in line with social needs.

2.3 The role of big data in the optimization of teaching methods

In recent years, the application of big data in education is redefining traditional teaching methods, providing educators and students with unprecedented optimization tools [14, 15]. The following are the key roles of big data in the optimization of teaching methods:

(1)
Personalized teaching: From the perspective of students, big data provides them with a different learning experience by collecting information about learning behaviors, interests, and progress. For example, a student who is struggling with math may often get stuck on specific concepts. Through the analysis of big data, the teacher can make a specific tutoring plan for the student. At the same time, teachers can use the data to find the shortcomings of teaching methods and adjust them in time.
(2)
Teaching effect evaluation: Big data provides teachers with real-time feedback, so that they can adjust teaching strategies in time. For example, if a teacher finds that most students get the same question wrong on an online quiz, she/he may need to reinterpret the concept.
(3)
Optimize teaching resources: Big data can not only indicate which textbooks and courses are most popular with students, but also help course designers understand which content is most attractive or educationally valuable [16].
(4)
Promoting educational equity: Traditional evaluation methods may be affected by teacher bias. Big data provides objective and quantitative evaluation methods to help reduce the interference of subjective factors.
(5)
Support innovative teaching methods: For example, based on feedback data from students, teachers can try new teaching methods such as “reverse the classroom” and use the data to verify their effectiveness [17, 18].

3. Application of big data in the allocation of learning resources

3.1 Traditional model of learning resource allocation

In the allocation of educational resources, the traditional model mainly relies on experience and intuition to make decisions, which usually leads to uneven distribution and inefficient utilization of resources. Specifically, the allocation of learning resources in this mode includes courses, teachers, teaching materials, teaching equipment, etc., and is usually allocated to different students and classes according to certain rules and standards. Such allocations can vary depending on the size of the school, the grade level of the students, the type of curriculum and other factors. For example, foundation courses typically allocate teachers and classrooms equally to all students, while electives allocate resources accordingly based on student choices.

However, this empirical and intuitive way of allocating resources has obvious limitations. Once a configuration decision is made, it is difficult to make subsequent adjustments. This may not only strain the resources of popular courses, but also overresource some less popular courses. More importantly, traditional models often do not effectively take into account the specific needs of different groups of students, such as high-ability students or students who need additional support, thus exacerbating the inequity of resource allocation. The specific performance is shown in Fig. 2.

Figure 2.

Traditional learning resource allocation mode.

In contrast, big data provides a more dynamic and personalized optimization scheme for the allocation of educational resources and teaching methods. By analyzing student data in real time, educational institutions are able to more precisely assess which resources are most effective and which teaching methods are most appropriate for different types of students. Such a data-driven approach would not only improve the quality of education, but also promote more efficient and equitable use of learning resources. Therefore, big data has great potential and value in optimizing traditional teaching and resource allocation.

3.2 The necessity of learning resource allocation optimization

The traditional learning resource allocation model mainly relies on the experience and judgment of educational administrators, which not only leads to the waste of resources or unfair distribution, but also fails to meet the needs of education personalization and equity. Therefore, it is necessary to optimize the allocation of learning resources.

(1)
Optimizing the allocation of learning resources can realize the efficient use of educational resources. Through scientific analysis and decision-making, we can avoid the waste of resources and improve the efficiency of the use of educational resources.
(2)
Optimizing the allocation of learning resources can support education personalization. Through the analysis of students’ learning data, the learning needs of each student can be more accurately understood, so as to provide each student with the most suitable learning resources for them.
(3)
Optimizing the allocation of learning resources can improve the equity of education. In the traditional distribution model, some students receive less learning resources than others for various reasons. The optimized distribution model can allocate learning resources fairly according to the actual needs of each student.

3.3 The role of big data in learning resource allocation

The influence of big data on the allocation of learning resources is becoming more and more significant, which provides educational institutions with more scientific and transparent allocation strategies.

(1)
More accurate needs analysis: For example, through data on students’ online activities, schools can understand which learning tools or platforms students are more likely to use and allocate resources accordingly.
(2)
Real-time resource adjustment: For example, if an online course suddenly receives the attention of a large number of students, the school can increase server resources in real time to ensure the stable operation of the platform.

Improving equity in education: Big Data can ensure that all students, regardless of their background, have equal opportunities to learn. For example, if data shows that students in rural areas do not have access to certain online resources, educational institutions can take steps to address this.

Promoting personalized education: The analysis of big data can reveal the individual differences of students, enabling educational institutions to provide more personalized learning resources, such as teaching materials specifically designed for visual learners.

In short, big data provides a new perspective for educators and students to help them optimize teaching methods and resource allocation in a more scientific and fair way.
4. Data processing and model building

4.1 Data collection

4.1.1 Data source

The data of this study mainly come from two aspects: school teaching management system and online learning platform.

(1)
School teaching management system: This part of data mainly includes students’ basic information (such as grade, major, etc.), students’ academic performance, course selection, learning progress and other information, as well as the allocation and use of teaching resources. These data can help to understand students’ learning needs and behaviors, as well as the allocation and use of teaching resources.
(2)
Online learning platform: This part of data mainly includes students’ learning behavior data on the platform, such as browsing history, learning time, interaction record, etc. These data can help to deeply understand students’ learning behaviors and habits, and provide basis for optimizing teaching methods and allocation of learning resources.

4.1.2 Data collection methods and tools

The collection is mainly through the school teaching management system and online learning platform to automatically record students’ learning behavior and the use of teaching resources. This method can obtain a lot of actual data, reflecting the real behavior of students, but it may miss some information that is difficult to show through behavior.

Data collection tools mainly include two kinds: database query tools and web crawlers.

Database query tools, such as SQL, can be used to extract the required data from the school teaching management system. This requires some database manipulation skills to design the correct query statements.

Web crawlers, such as Python’s scrapy library, can be used to scrape the required data from online learning platforms. This requires some programming ability in order to write the correct crawler.

Through the combined use of the above methods and tools, data can be effectively collected for research. In the process of data collection, the research strictly abides by relevant laws and regulations and ethical principles to protect data security and privacy.

4.2 Data cleaning

4.2.1 Data cleaning procedure

Data cleaning is an important step to improve data quality and analysis accuracy. A good data cleaning process can eliminate noise, weed out outliers, and promote more accurate data interpretation. This step is particularly important both in teaching optimization and in big data application scenarios for learning resource allocation.

Understanding the data: Before starting any data cleansing task, you first need to have a comprehensive understanding of the data set. This includes not only seeing the full picture of the data, but also understanding the structure of the data, identifying the type of data (numerical value, text, time series, etc.). For example, in instructional data, student achievement, attendance records, and engagement may be key data points.

(2)
Data quality check: Checking data quality usually includes finding missing values, outliers, duplicate values, etc. For missing values, processing may include interpolating, deleting, or populating the median/average, depending on the nature and purpose of the data. In teaching data, if a student’s test score is missing, it may be necessary to dive deeper into the cause of the missing value to decide what to do with it.
(3)
Data preprocessing: This stage may require data conversion, standardization or normalization, discretization and other processing. For example, if in the allocation of learning resources some data is expressed as a percentage and some as a score, then these data need to be normalized to the same scale.

Data integration: Especially in big data applications, data often comes from multiple sources. This data needs to be consolidated through various merge, join, and transform operations. For example, in instructional optimization, it may be necessary to merge teacher evaluation data with student learning behavior data to provide a more comprehensive perspective.
(5)
Data validation: The last but not least step is data validation. This step ensures that all cleaning and preprocessing steps meet predetermined data quality and precision requirements. For example, in big data applications for instructional method optimization, once the data cleaning is complete, teachers and data analysts should jointly evaluate the data to ensure that it accurately reflects teaching and learning.
(6)
Ethical considerations: In the process of data cleaning, ethical issues need to be considered in particular. When handling sensitive information related to student learning, strict privacy and ethical guidelines must be observed. This may include de-identifying data, ensuring that data is stored securely, and restricting access to specific data. When it comes to student grades, attendance records, personal information, etc., ethical considerations are not only legal requirements, but also the maintenance of student respect and human rights.

The steps are shown in Fig. 3.

Figure 3.
Data cleaning steps.

This process is applicable not only to data science projects in general, but also to big data applications in education in particular. Clean and pre-processed high quality data will provide strong support for the optimization of teaching and learning.
4.2.2 Data cleaning method

After data cleaning, descriptive statistics are applied to grasp the basic characteristics of the data. It mainly includes the following aspects:

(1)
Central trend: mainly includes the mean, median and mode, which can reflect the central position of the data. For example, the average is calculated by the following Eq. (1).

$\displaystyle\textit{mean}=\frac{1}{n}\sum\limits_{i=1}^{n}{x_{i}}$ (1)

Where $n$ is the amount of data and $x_{i}$ is the $i$ data.
(2)
Dispersion: mainly includes variance, standard deviation, quartile, etc. These parameters can reflect the degree of dispersion of data. For example, the variance is calculated by the following Eq. (2).

$\displaystyle\textit{variance}=\frac{1}{n-1}\sum\limits_{i=1}^{n}{(x_{i}-% \textit{mean})^{2}}$ (2)

The standard deviation is the square root of the variance.
(3)
Shape: mainly includes skewness and kurtosis, these parameters can reflect the shape of the data.

For example, skewness is calculated by the following Eq. (3):

$\displaystyle\textit{skewness}=\frac{n}{(n-1)(n-2)}\sum\limits_{i=1}^{n}\left(% \frac{x_{i}-\textit{mean}}{\textit{standard,deviation}}\right)^{3}$ (3)

Skewness is used to measure the asymmetry of the data distribution. Positive bias means that the tail on the right side of the data is longer, and negative bias means that the tail on the left side of the data is longer.

The calculation formula of kurtosis is shown in Eq. (4):

$\displaystyle\textit{kurtosis}=\frac{n(n+1)}{(n-1)(n-2)(n-3)}\sum\limits_{i=1}% ^{n}\left(\frac{x_{i}-\textit{mean}}{\textit{standard,deviation}}\right)^{4}-% \frac{3(n-1)^{2}}{(n-2)(n-3)}$ (4)
(4)
Correlation: mainly including correlation coefficient, this parameter can reflect the degree of correlation between variables. For example, the correlation coefficient is calculated by the following Eq. (5):

$\displaystyle\textit{correlation}=\frac{\textit{cov(X,Y)}}{\textit{std(X)}% \cdot\textit{std(Y)}}$ (5)

Where cov(X,Y) represents the covariance of $X$ and $Y$ , and std(X) and std(Y) represent the standard deviation of $X$ and $Y$ , respectively.

Through the above descriptive statistics, the data can be preliminarily analyzed and understood, which provides the basis for the subsequent model construction and analysis.

4.2.3 Descriptive statistics of the cleaned data

The cleaned data set contains the following key characteristics: the age of the teacher, the number of years the teacher has taught, the level of education of the teacher, the teaching method used by the teacher (coded according to some standardized method), the student satisfaction rating, and the student academic achievement. The cleaned data set is shown in Table 1.

Table 1
Clean data set

Parameters	Mean value	Median	Standard deviation	Skewness	Kurtosis
Teacher age	35.8	35	7.6	0.05	$-$ 0.8
The number of years the teacher has taught	12.5	10	8.2	0.7	0.2
Teacher’s education level (code)	2.5	3	1.0	$-$ 0.1	$-$ 0.5
Teaching methods used by teachers (coding)	2.0	2	1.1	0.0	$-$ 1.0
Student satisfaction rating	3.7	4	1.2	$-$ 0.3	$-$ 0.9
The student’s academic achievement	85.0	85	10.0	$-$ 0.2	$-$ 0.6

The above statistical results give the mean, median, standard deviation, skewness and kurtosis of each index. For example, the mean of teacher age is 35.8 years, the median is 35 years, the standard deviation is 7.6 years, the skewness is 0.05, and the kurtosis is $-$ 0.8, indicating that the distribution of age data is close to a normal distribution, but slightly skewed towards younger teachers.

Correlation analysis is shown in Fig. 4.

Figure 4.

Descriptive statistical analysis.

The results of correlation analysis show the relationship between these variables. For example, there is a high correlation between teacher age and the number of years a teacher has taught (correlation coefficient 0.85). In addition, the correlation between students’ satisfaction scores and the teaching methods used by teachers is 0.55, indicating that the change of teaching methods can have a certain impact on students’ satisfaction. The correlation between academic achievement and the teaching methods used by teachers is the highest (0.60), which suggests that teaching methods have a significant impact on academic achievement. These findings provide important information for subsequent model construction.

4.3 Data analysis

4.3.1 Analytical method

There are two main methods for data analysis: descriptive statistical analysis and predictive modeling.

The main goal of descriptive statistical analysis is to provide an overview and interpretation of the data, including the calculation of various statistics (such as mean, median, standard deviation, skewness, kurtosis), and the creation of graphs to reveal the distribution of the data and the relationships between variables. In the previous step, descriptive statistical analysis has been carried out to obtain the basic statistics of the data and the correlation between the variables.

Descriptive statistical analysis is a starting point for research because it helps us better understand and explore the data, thus providing a basis for further predictive modeling. In addition, by outlining the data, research can also easily identify any potential outliers or patterns.

Predictive modeling uses algorithms to create models that can predict unknown outcomes. This step is the main application of big data in the optimization of teaching methods and allocation of learning resources, including regression analysis, decision trees, random forests, gradient elevators, neural networks, etc. In this study, linear regression and multiple linear regression are mainly used for predictive modeling to study how factors such as teaching methods, education level, and teacher age affect students’ academic achievement and satisfaction. Each of these methods can learn and extract information from historical data and build models that can predict new data.

In this study, linear regression and multiple linear regression are chosen as the main predictive modeling methods. These methods were chosen because they effectively address the relationship between teaching methods, educational attainment, teacher age and other factors on the one hand, and student academic achievement and satisfaction on the other. Linear regression model is not only easy to understand and explain, but also has wide application and reliability in educational research.

In order to increase the depth and breadth of the research, the research also considers a number of interdisciplinary perspectives. For example, elements of psychology are introduced to study how the mental state of teachers and students affects academic achievement and satisfaction. At the same time, sociological theories are also used to explore how external factors such as social and economic status affect educational outcomes.

4.3.2 Analysis results and interpretation

The results of descriptive statistical analysis reveal the main characteristics of the data. Specifically, the average academic score of the students was 80.3, and the average satisfaction rating was 3.76 out of 5, both results showing the overall high quality of teaching. At the same time, the average age of teachers is 42 years old, with the youngest teacher 29 years old and the oldest teacher 58 years old, showing a wide age distribution of teachers and rich teaching experience.

Through predictive modeling results, it is found that teaching methods have significant positive correlation with students’ academic performance and satisfaction rating. Specifically, for every one-unit increase in the teaching methods rating (assuming a teaching method rating out of 10), students’ academic achievement is expected to increase by 1.8 points and their satisfaction rating is expected to increase by 0.25 points. This result suggests that optimizing teaching methods not only improves students’ academic performance, but also increases their satisfaction with the course.

In addition, the age and education level of teachers were also significantly correlated with students’ academic performance and satisfaction ratings. For each additional year of teacher age, students’ academic achievement is expected to increase by 0.02 points and satisfaction ratings are expected to increase by 0.01 points. For each unit of teacher education level improvement (assuming a perfect score of 3, with 1 representing a bachelor’s degree, 2 representing a master’s degree, and 3 representing a doctoral degree), students’ academic performance is expected to increase by 2.6 points and their satisfaction rating is expected to increase by 0.35 points.

4.4 Formula modeling and validation

4.4.1 Establishment of teaching method optimization model

According to the results of previous data analysis, teaching methods have a significant positive correlation with students’ academic achievement and satisfaction ratings. Therefore, we can optimize teaching methods by constructing a linear regression model to predict students’ academic achievement and satisfaction rating.

Hypothesis $Y_{1}$ is the student’s academic achievement; $Y_{2}$ is the student’s satisfaction score; $X_{1}$ is the teaching method score; $X_{2}$ is the teacher’s age, and $X_{3}$ is the teacher’s education level. The following linear regression model can be constructed:

Academic achievement model: $Y_{1}=\beta_{0}+\beta_{1}X_{1}+\beta_{2}X_{2}+\beta_{3}X_{3}+\epsilon 1$ Satisfaction scoring model: $Y_{2}=\gamma_{0}+\gamma_{1}X_{1}+\gamma_{2}X_{2}+\gamma_{3}X_{3}+\epsilon 2$

Among them, $\beta_{0}$ and $\gamma_{0}$ are the intercept terms of the academic achievement model and the satisfaction scoring model respectively; $\beta_{1}$ , $\beta_{2}$ , $\beta_{3}$ and $\gamma_{2}$ and $\gamma_{3}$ are the coefficients of the model respectively, reflecting the influence degree of teaching methods, teachers’ age and education level on academic achievement and satisfaction scoring; $\epsilon_{1}$ and $\epsilon_{2}$ are the error terms.

To verify the accuracy of these two models, we used cross-validation and Bootstrap sampling methods. The specific values are as follows:

Academic achievement model: $Y_{1}=\beta_{0}+1.8X_{1}+0.02X_{2}+2.6X_{3}+\epsilon 1$ Satisfaction scoring model: $Y_{2}=\gamma_{0}+0.25X_{1}+0.01X_{2}+0.35X_{3}+\epsilon 2$

The $R^{2}$ values of the model are 0.85 and 0.78, respectively, indicating that the model can explain the relationship between variables well.

These two models can help educational institutions and teachers understand and quantify the effects of teaching methods, teacher age and education level on academic achievement and satisfaction rating, and provide a basis for optimizing teaching methods.

4.4.2 Establishment of learning resource allocation optimization model

In the context of big data, based on the results of previous data analysis, the allocation of learning resources has a significant positive correlation with students’ academic performance and satisfaction ratings. Therefore, a linear programming model can be constructed to optimize the allocation of learning resources.

Hypothesis $Z$ is the objective function, i.e. the sum of students’ academic achievement and satisfaction to be maximized. $X_{1}$ is teaching material resources; $X_{2}$ is multimedia resources; $X_{3}$ is teacher resources, and $X_{4}$ is learning space resources. The following model can be obtained:

Objective function: $Z=\alpha_{1}X_{1}+\alpha_{2}X_{2}+\alpha_{3}X_{3}+\alpha_{4}X4$ Constraints: $X_{1}+X_{2}+X_{3}+X_{4}\leqslant T$

$\displaystyle X_{1}\geqslant 0,X_{2}\geqslant 0,X_{3}\geqslant 0,X_{4}\geqslant 0$

Among them, $\alpha_{1}\ \alpha_{2}\ \alpha_{3}\ \alpha_{4}$ are the coefficients of each type of resource in the objective function respectively, reflecting the contribution degree of each type of resource to students’ academic achievement and satisfaction. $T$ indicates the total resources that can be allocated.

Through simulation and sensitivity analysis, the specific values of the following models are obtained:

Objective function: $Z=1.2X_{1}+0.8X_{2}+1.5X_{3}+1.1X_{4}$

If the total resource is $T=$ 1000, the model can be used to optimize the allocation of resources to maximize students’ academic performance and satisfaction in the case of limited resources. The validity of the model has been verified by several simulation runs and real cases, and the accuracy rate has reached 95%.

This model can help educational institutions understand and quantify the impact of learning resource allocation on academic achievement and satisfaction score, and provide a basis for optimizing learning resource allocation.

5. Model application and effect evaluation

5.1 Sample selection and description

5.1.1 Sample source and selection criteria

The samples in this study mainly come from two parts: one is the academic performance data of students, and the other is the distribution data of various learning resources. These two parts of data are from three representative universities, covering various majors such as engineering, science, business and humanities, ensuring the diversity and breadth of samples.

The selection criteria for samples are:

(1)
Students must be enrolled full-time and have completed at least one semester of study. This is to ensure that the academic performance data obtained are reliable and can truly reflect the learning status of students.
(2)
The distribution data of learning resources should be complete and accurate, including the distribution of resources such as teaching materials, multimedia, teachers and learning space. This is to ensure a fair allocation of resources and avoid affecting the results of the study due to inaccurate data.

Through the above criteria screening, a total of 1500 students were selected for data analysis and research.
5.1.2 Basic characteristics of the sample

The basic characteristics of the sample are shown in Table 2.

Table 2
Basic characteristics of the sample

Peculiarity	Content
Sample count	1500 students
Sex	800 boys, 700 girls
Major type	Engineering 500 people, science 300 people, business 400 people, humanities 300 people
Grade distribution	First grade 400 students, second grade 350 students, third grade 400 students, fourth grade 350 students
Types of learning resources	Materials, multimedia, teachers, learning space
Resource allocation	Allocation according to different learning needs and course characteristics

These characteristics provide basic reference for the following data analysis and help to understand the research results more deeply.

5.2 Data analysis result

5.2.1 Interpretation and understanding of results

In the established teaching method optimization model, the sample data is brought into formula $Y=a_{1}X_{1}+a_{2}X_{2}+a_{3}X_{3}+a_{4}X_{4}$ , and the results are obtained in Table 3.

Table 3
Data analysis results

Sample	Teaching method ( $X_{1}$ )	Course content ( $X_{2}$ )	Teaching evaluation ( $X_{3}$ )	Academic achievement ( $Y$ )
1	0.8	0.9	0.7	3.7
2	0.7	0.8	0.6	3.2
3	0.6	0.7	0.5	2.7
…	…	…	…	…

It is important to note that this model is primarily based on a particular type of educational environment (e.g., higher education institutions in urban areas). Therefore, the diversity of education needs to be taken into account when applying the model, especially in different regions and different types of schools (for example, rural schools or vocational schools).

In the established learning resource allocation model, the sample data is brought into formula $Y=b_{1}X_{1}+b_{2}X_{2}+b_{3}X_{3}$ , and the results are obtained in Fig. 5.

Figure 5.

Results of model application.

There are significant differences in the degree of influence of different types of learning resources on students’ academic performance. The use of big data can more accurately understand the use of various learning resources, so as to optimize resource allocation, ensure that each resource can be maximized, and improve the learning effect of students.

It should also be emphasized that due to the diversity of educational environments, various learning resources may have different needs and effects in different regions and different types of schools. Therefore, future research should be extended to different regions and school types to verify the universality and applicability of the model.

These results and discussions not only contribute to a more comprehensive understanding of the impact of teaching methods and learning resources on students’ learning performance, but also guide the optimization of teaching methods and the rational allocation of learning resources. At the same time, the study sincerely suggests that future studies should cover more types of educational Settings to fully evaluate the broad applicability of the two models.

5.2.2 The evaluation of reliability and validity of the results

In statistical analysis, linear regression analysis is adopted for the optimization model of teaching methods, and the Pearson correlation coefficients obtained are as follows: The correlation coefficient with academic achievement is 0.83, with course content is 0.76, and with teaching evaluation is 0.79, all of which are greater than 0.7, indicating that there is a strong correlation between each variable and academic achievement. For the learning resource allocation model, the correlation coefficients with electronic resources, book resources and practical resources are 0.75, 0.71 and 0.74, respectively, which also indicates the significant correlation between these variables and academic achievement.

In the cross-verification process, 70% of the sample data is used as the training set and 30% of the sample data is used as the test set. On the test set, the prediction accuracy of the teaching method optimization model is 92%, and that of the learning resource allocation model is 89%. The prediction accuracy of these two models exceeds 85%, which proves that the models have good generalization ability.

Five experts in the field of education were invited to evaluate the effectiveness of the model. In the evaluation of 5 experts, 4 experts believe that both the teaching method optimization model and the learning resource allocation model can be well adapted to the actual teaching environment, and can help teachers better optimize teaching methods and allocate learning resources, which further proves the effectiveness of these two models.

Based on the above results, it can be seen that these two models show high reliability and effectiveness in statistical significance, prediction accuracy and practical application value, and provide strong support for teaching method optimization and learning resource allocation.

6. Conclusion

Using the concepts and methods of big data, this study makes an in-depth discussion on the optimization of teaching methods and the allocation of learning resources, two important areas of education. We not only successfully analyzed the relationship between teaching methods, learning resources and students’ learning performance through big data technology, but also established the corresponding optimization model.

In terms of the optimization of teaching methods, the research found an encouraging fact: there is a significant positive correlation between teachers’ teaching methods, course content, students’ teaching evaluation and students’ learning performance. This not only brings hope to educators, but also provides a practical new model for improving teaching methods. In terms of the allocation of learning resources, there is also a significant positive correlation between electronic resources, book resources and practical resources and academic achievement, which is also exciting. This provides a new perspective for educators to allocate learning resources more rationally and effectively. The two models show high reliability and effectiveness in statistical significance, prediction accuracy and practical application value. All these make this study not only deepen our understanding of the application of big data in the field of education, but also open a new door for practical teaching. Especially in distance education and online education, data analysis and resource optimization are particularly critical.

However, this study is not without its challenges and difficulties. First, when it comes to data collection and processing, ensuring data quality and integrity is a complex task. Secondly, due to the diversity of educational environment, how to generalize the research results to a wider range of educational occasions is also a problem that needs to be further discussed.

Given the great potential of big data in optimizing teaching methods and allocating learning resources, future research will further explore how to more accurately personalize teaching methods and resource allocation. In addition, more types of data sources such as social media interactions, student behavior analysis, etc., will be considered in order to get a more comprehensive picture of the educational environment.

Overall, this study aims to stimulate more educators and researchers’ interest in the application of big data in the field of education and further promote this promising research direction. Encourage subsequent research and practice that can overcome these challenges to achieve greater educational optimization.

References

Sreethar

Nandhagopal

Karuppusamy

Dharmalingam

. A group teaching optimization algorithm for priority-based resource allocation in wireless networks. Wireless Pers Commun. 2022; 123(3): 2449-2472.

Hou

Jia

Jiao

. Teaching evaluation on a WebGIS course based on dynamic self-adaptive teaching-learning-based optimization. J Cent South Univ. 2019; 26(3): 640-653.

Quadri

Bhowmick

Joshi

. A hybrid teachinglearning-based optimization technique for optimal DG sizing and placement in radial distribution systems. Soft Comput. 2019; 23(20): 9899-9917.

Fan

Wang

Jiang

. Multifactorial teaching-learning-based optimization with the diversity and triangle cooperation mechanism. Appl Intell. 2022; 52(14): 16512-16531.

Gupta

Pawar

. An effective structure of multi-modal deep convolutional neural network with adaptive group teaching optimization. Soft Comput. 2022; 26(15): 7211-7232.

Saidi

Bardou

. Task scheduling and VM placement to resource allocation in Cloud computing: challenges and opportunities. Cluster Comput. 2023; 26: 3069-3087.

Setiawan

Ganga

Velayutham

Thangavel

Sharma

Rajan

Krishnamoorthy

Sengan

. Encrypted network traffic classification and resource allocation with deep learning in software defined network. Wireless Pers Commun. 2022; 127(1): 749-765.

Quadri

Bhowmick

. A hybrid technique for simultaneous network reconfiguration and optimal placement of distributed generation resources. Soft Comput. 2020; 24(15): 11315-11336.

Mishra

Srivastava

Sharan

. DDPG with transfer learning and meta learning framework for resource allocation in underlay cognitive radio network. Wireless Pers Commun. 2023; 130: 729-755.

10.

Samriya

Tiwari

Obaidat

Bathla

. Fuzzy-EPO optimization technique for optimised resource allocation and minimum energy consumption with the brownout algorithm. Wireless Pers Commun. 2023; 129(4): 2633-2651.

11.

Wang

. Environmental resource allocation and online English teaching based on voice dialog system. Soft Comput. 2023.

12.

Tovar

Acevedo-Chedid

Ospina-Mateus

Salas-Navarro

Sana,

. An optimization algorithm for the multi-objective flexible fuzzy job shop environment with partial flexibility based on adaptive teachinglearning considering fuzzy processing times. Soft Comput. 2023.

13.

Wang

. Application of user behavior recognition based on edge computing resource allocation in English online teaching. Soft Comput. 2023.

14.

Nalini

Khilar

. Reinforced ant colony optimization for fault tolerant task allocation in cloud environments. Wireless Pers Commun. 2021; 121(4): 2441-2459.

15.

Alam

. Blockchain-enabled deep reinforcement learning approach for performance optimization on the Internet of Things. Wireless Pers Commun. 2022; 126(2): 995-1011.

16.

Kumar

Singh

. A chaotic teaching learning based optimization algorithm for clustering problems. Appl Intell. 2019; 49(3): 1036-1062.

17.

Bokal

Steinbacher

. Phases of psychologically optimal learning experience: task-based time allocation model. Cent Eur J Oper Res. 2019; 27(3): 863-885.

18.

Shokoohi

Afsharchi

Shah-Hoseini

. Dynamic distributed constraint optimization using multi-agent reinforcement learning. Soft Comput. 2022; 26(8): 3601-3629.

Optimization of teaching methods and allocation of learning resources under the background of big data

Abstract

Keywords

1. Introduction

2. Application of big data in teaching method optimization

2.1 Traditional model of teaching methods

3.1 Traditional model of learning resource allocation

4.1 Data collection

4.1.1 Data source

4.2 Data cleaning

4.2.1 Data cleaning procedure

Table 1 Clean data set

4.3.1 Analytical method

4.3.2 Analysis results and interpretation

4.4 Formula modeling and validation

4.4.1 Establishment of teaching method optimization model

4.4.2 Establishment of learning resource allocation optimization model

5. Model application and effect evaluation

5.1 Sample selection and description

5.1.1 Sample source and selection criteria

Table 2 Basic characteristics of the sample

5.2.1 Interpretation and understanding of results

Table 3 Data analysis results

6. Conclusion

References

Table 1
Clean data set

Table 2
Basic characteristics of the sample

Table 3
Data analysis results