Application of English education big data system based on multi-source information fusion and machine learning

doi:10.21203/rs.3.rs-2716102/v1

Download PDF

Research Article

Application of English education big data system based on multi-source information fusion and machine learning

https://doi.org/10.21203/rs.3.rs-2716102/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 22 May, 2023

Read the published version in Soft Computing →

You are reading this latest preprint version

For administrators in the field of education, the data generated in the actual education process is difficult to achieve effective management due to its large amount of information, complex structure and multiple types. However, in the actual construction of education informatization, it is impossible to explore the laws and development trends of education information only by manual analysis, so it is difficult to obtain effective concepts and relevant views. College education administrators can not fully optimize the process of teaching diagnosis and reform through education informatization. Based on this background, this research introduces machine learning technology to achieve technical support for the integrated application of educational data. This paper focuses on the design and construction of an English teaching management data analysis system for college teaching, which combines machine learning technology with multi-source information technology. The system is mainly composed of hardware layer, data layer, service layer and application layer, which can assist teaching and realize learning career planning, faculty authority management, curriculum data modification, teaching resource allocation and scientific research review. The simulation results show that this big data system has high processing efficiency, and is superior to other algorithms in terms of mean value and FVDF extreme value, so it can effectively assist English teaching management. This paper completes the design of English education big data system by introducing machine learning and multi-source information fusion into the field of education.

multi-source information fusion

Machine learning

English education

Big data system

The importance of English subject has been highlighted with the continuous development of social economy. Economic globalization has enabled countries around the world to interact frequently [1]. As the official language of the world, the importance of English is self-evident. It is not only learned as a language or a way of communication, but also widely used in various fields and has become one of the most important skills in the current world information exchange environment [2]. Therefore, most countries have set up English teaching courses and incorporated them into the basic education curriculum system of schools to improve the national English level. English education is not only a part of quality education, but also an important channel to reflect students' communication skills [3]. In China, English is one of the compulsory subjects for primary and secondary schools, universities and graduate students. In the college entrance examination, CET 4 and CET 6 and the entrance examination for graduate students, English scores account for a considerable proportion [4]. Therefore, how to learn English well and how to communicate in English become more and more important. In recent years, with the development of the Internet, great changes have taken place in all sectors of society [5]. "Internet + education" has gradually become a development trend, and online education platforms have attracted more and more attention. In this context, students are more willing to learn through the online English learning platform, and schools are also willing to use the online education system to complete teaching [6]. In the actual teaching process of college English, teachers can teach students to listen, speak, read and write English through the online English platform, making the learning process of students easier and more effective.

Based on the teaching practice and the process and technical characteristics of data mining, the literature proposes an intelligent teaching system model based on data mining, and then analyzes the possibility of history mining and the use of association rules in data analysis during students' learning [7]. An approximate association analysis algorithm based on principal attributes and low-frequency rules is proposed to analyze the association rules between course contents. According to the basic information characteristics of students, the classic ID3 decision tree algorithm is used to classify students [8]. Combined with the results of the analysis of English teaching needs, the literature gives the system use case diagram, discusses the design of the whole system, including the design of the development environment, the design of the operating environment, the design of the functional modules, and provides a simple flow design diagram [9]. Then, the overall E-R diagram of the system is drawn, and all the logical table structures of the system are designed. Literature has developed a set of online learning system based on Microsoft SQL Server 2000 and ASP technology, which corresponds to the current college English teaching needs [10]. The system is divided into three modules: system management, student learning and teachers. Then the design and implementation of each sub module of teacher management and student learning are introduced in detail. Finally, the system is tested and evaluated [11]. The system has achieved the expected design purpose, and the rules obtained through data mining can realize the basic management of student information, teaching plans, exercises and online tests, as well as the learning and testing functions of online courses for students at different levels [12]. Appropriate and targeted method suggestions can be provided. In the literature, the clustering criteria were optimized by using an improved differential evolution algorithm. The algorithm is used to analyze and calculate the learning content of students' English courses, form intuitive rules for students' learning records and interests, and help students make personalized learning plans [13]. The literature uses the enhanced association rule algorithm to sort the information accessed by students, identify high-frequency web pages accessed by students, and identify the intuitive location of high-frequency access and learning disabilities [14]. It is helpful for schools to make educational decisions, find problems in students' English learning process, improve the content of online teaching platform, and complete the task of personalized online education.

Linear regression (LR) multiplies each feature or attribute of the training sample set by a parameter, and adds the results as the output of the results. It is expressed as:

$$\text{h}\left(\text{x}\right)={\text{w}}^{\text{T}}\text{x}$$

Logical regression is to put the sum obtained from linear regression into the sigmoid function and get the decimal between (0, 1). If the value is greater than the specified threshold, it is considered as a positive class, and if it is less than the threshold, it is considered as a negative class, namely:

$$\text{g}\left(\text{z}\right)=\frac{1}{1+{\text{e}}^{-\text{w}\text{x}}}$$

Suppose there are n training samples, the probability of each sample occurrence corresponds to the Bernoulli distribution, and p (yi = 1/x) represents the probability of a positive class, then the probability of a negative class is 1-p (yi = 1/x), expand it to solve it, and take the derivative of w:

$$\frac{\partial \text{l}\left(\text{w}\right)}{\partial \text{w}}=\sum _{\text{i}=1}^{\text{m}} \left({\text{y}}_{\text{i}}-\text{g}\left(\text{z}\right)\right){\text{x}}_{\text{i}}$$

The support and confidence formula is described as follows:

$$\text{support }(\text{C}\Rightarrow \text{D})=\left|\text{C}\text{U}\text{D}\right|/\left|\text{E}\right|$$

$$\text{confidence }(\text{C}\Rightarrow \text{D})=\left|\text{C}\text{U}\text{D}\right|/\text{C}\mid$$

Set thresholds for support and confidence. Strong rules are rules that meet support and confidence criteria. The association rule analysis process can be briefly described as follows:

$$\text{e}\text{n}\text{t}\text{r}\text{o}\text{p}\text{y}\left(\text{A}\right)=-\sum _{\text{i}=0}^{\text{m}} {\text{p}}_{\text{i}}{\text{l}\text{o}\text{g}}_{2}{\text{p}}_{\text{i}}$$

If attribute E divides training dataset A into n pieces, Ai is the ith subset of the training dataset, and | A | and | Ai | are the number of training dataset A and its subset Ai, and then divide these subsets. Subset entropy, the expected value formula can be expressed as:

$$\text{e}\text{n}\text{t}\text{r}\text{o}\text{p}\text{y}(\text{A},\text{E})=\sum _{\text{i}=0}^{\text{m}} \frac{\left|\text{A}\text{i}\right|}{\left|\text{A}\right|}\text{ entropy }\left({\text{A}}_{\text{i}}\right)$$

Then the training data set A and attribute E are divided, and the information obtained by the entropy reduction value is the difference between the undivided entropy value and the segmented entropy value, gain (A, E) = entropy (A) - entropy (A, E). The higher the information obtained, the more classification information provided by the selected partition attribute, because entropy is a measure of data diversity. If less information is obtained, it indicates that the classification attribute lacks diversity. The classification information provided by this attribute is very little and has no value as a classification attribute.

Figure 1 shows the difference in execution time between the optimized approximate association analysis algorithm and the classic Apriori association analysis algorithm when setting the minimum support and the number of transaction records in the database. It can be seen from the figure that when the transaction record is large enough, the efficiency of the optimization algorithm is higher than that of the classic Apriori algorithm, which shows the efficiency and availability of the algorithm.

Through clustering experiments on two datasets, the number of iterations and convergence results are shown in Fig. 2. After comparing the results of different clustering algorithms, it is found that the F-measure value of standard differential evolution algorithm and improved differential evolution algorithm is higher than that of K-means algorithm, whether it is Iris dataset or Wine dataset. It can be proved that K-means algorithm is a relatively poor algorithm and its precision is lower than the first two algorithms.

The fitness function is shown in Table 1.

Table 1

Comparison of fitness functions of different algorithms on two datasets
Data set	Iris			Wine
Condition	mean value	Worst value	optimal value	mean value	Worst value	optimal value
GA	126.44	141.18	115.12	16695.84	16695.84	16695.84
TS	98.84	99.56	98.33	16953.31	17005.92	16832.88
SA	98.10	98.23	98.07	16695.84	16695.84	16695.84
ACO	98.14	98.79	98.07	16695.84	16695.84	16695.84
DECluster	97.62	97.62	97.62	16455.14	16455.14	16455.14
IDECluster	97.62	97.62	97.62	16455.11	16455.11	16455.11

4.1 System frame structure design

This English teaching management data analysis system adopts a conventional system development architecture. The platform development architecture is shown in Fig. 3. The whole system consists of four layers: hardware layer, data layer, service layer and application layer.

4.2 System function design

The system function design diagram is shown in Fig. 4.

The visualization display page is the key function page of the entire visualization platform. It is mainly divided into five levels: schools, teachers, students, majors and courses. It is dedicated to viewing data processed in related heterogeneous systems. This page does not require login. School staff and visitors can access this page. Entering the login page through the visual display page is the entrance to the main subsystems such as the teacher development platform, student development platform, scientific research management system, and curriculum management system. Unified authentication, verifying user name and password, and obtaining corresponding user authorization. The subsystem can access without repeated authentication.

The teacher development platform mainly collects teachers' personal basic data, working data and planning data for the English teaching management data analysis system. In addition to data collection, the teacher development platform can also be a business platform. Different functions can be realized by dividing the functions of teachers, department heads, and personnel department heads. Teachers' authority is the basic authority. The workload is divided into three categories: teaching and professional development, scientific research and social services, and student management. The person in charge of the department gave feedback to the teacher to modify the authority, and the data collection was completed after the approval. The intelligent query function is mainly used by department heads to determine query conditions based on teachers' personal basic data or other data, to achieve multiple query conditions and easily find data results that meet the query requirements. The organization structure leader is the organization where the department head defines the school system structure. You can add or delete the organization head, or add or delete the teaching staff at different nodes of the organization.

4.3 System data simulation performance test

In order to avoid the influence of garbage data on the calculated results of spatial distribution entropy as much as possible, the data cleaning method is added to the calculated results of spatial distribution entropy of students in this model. The comparison is made through data cleaning, as shown in Table 2:

Table 2

Comparison of Student Spatial Distribution Entropy Calculation Results before and after Data Cleaning
School Name	Total number of students	Number of students with incorrect information	Number of students with valid information	Student spatial distribution entropy before cleanin	Student spatial distribution entropy after cleaning
class 1	666	8	658	0.871208	0.889845
class 2	1443	31	1412	0.831592	0.874827
class 3	1038	8	1030	0.862995	0.871675
class 4	1343	68	1275	0.803207	0.866185
class 5	848	41	807	0.823701	0.861907
class 6	1044	76	968	0.813542	0.854866
class 7	1455	103	1352	0.871208	0.889845

It can be seen from the comparison results that in the process of original data aggregation, incomplete information still exists in the data, leading to abnormal data in the statistical analysis database. Therefore, the entire model needs further optimization.

Table 3

Comparison of Task Throughput
Scheduling Algorithm	GA	SA	TS	ACO
Time period A	5866	808	895	5,400
Time period B	7,275	2,626	2,662	6,794
Time period C	7,811	4,444	4,627	7,484
Time period D	8,046	5,454	5,477	7,819
Time period E	8,222	5,858	5,953	8,000
Time period 6	8,306	6,262	6,327	8,087
Maximum value	2.94	0.91	0.98	2.70
Minimum	0.04	0.20	0.19	0.04
Average value	0.75	0.56	0.57	0.72

In order to evaluate the number of tasks completed in unit time, the experimenter measured the operation throughput and summarized the results through Table 3.

As shown in Fig. 5, it is the feature level fusion experiment results of DDoS attacks in the BPNN II model (that is, the second set of TIME, STATE, and PKTS features). In each generation, the average correct rate of the number of nodes listed is 3, 7, 9, and 11. The optimal model node number after optimization is 11, and the learning rate is 0.15.

5.1 Existing problems of education big data application

Big data application services are still at a disadvantage. The cloud service platform is still in its infancy. Big data service applications have not been popularized. There is a lack of professional technical teams to build and maintain the platform. Building and developing large-scale education platforms requires a large amount of data, and there is a lack of related services for later users.

Lack of political and financial support. The application of big data in the education industry is a developing field. At present, the application of education big data lacks political support and guidance, and the ownership of education big data and resources lacks legal norms. At the same time, in order to build an education platform for sharing big data in the city, there is a lack of special funds to support the construction and maintenance of the platform and the salaries of professionals.

The phenomenon of information island is serious. There are many educational resources in a city, and there are strong educational resources in universities, middle schools and primary schools. However, each university or library only sends educational resources to the database obtained by its own public platform. There is no unified platform to integrate all resources, and it is not open for sharing, which leads to the creation of information islands, and it is impossible to share all high-quality resources across regions. Therefore, the current situation of big data education application in a city cannot be trusted. The construction of urban education big data sharing platform should be supported and guided by the government and the competent education department.

5.2 Strategic design of education big data sharing platform

Build a big data sharing platform for urban education to help teachers and students better complete relevant teaching activities and promote personalized learning through data collection and analysis; At the same time, for education administrators, they can also find hidden and useful information from big data, which reflects the ability to improve education management through the education system. Therefore, this section will strategically design the urban big data sharing education platform from the aspects of framework structure, platform function, platform operation, etc.

To build an urban education big data sharing platform, it is necessary to establish a professional technical platform training and maintenance team under the support and guidance of the Municipal Education Commission, and follow the basic framework of building an education big data sharing platform. The system includes the base layer, the middle layer and the top layer.

Basic layer: In terms of hardware facilities, establish a shared network that can cover all networks, including educational resources of different universities, libraries and resource databases, so that the educational resources of various networks can penetrate each other and improve the interoperability between different network resources. At the same time, it improves the service capability of the terminal device, which can support various mobile devices and multimedia file operations in audio, video, software and other formats. In terms of software system, the architecture integrates three major software systems, namely IaaS, PaaS and SaaS, so that the urban construction education big data sharing platform can not only provide users with a variety of services, but also maintain the integrity of the collected education resources, ensure the security of all websites and platforms in the resource library related to the education big data sharing platform, and better serve the majority of users.

Middle layer: The education middle layer in the urban big data sharing platform is the core layer of the entire platform, which includes not only the collection of various data, but also the aggregation and storage of other various data, as well as the connection with various database resources. The input and output data can be controlled through the data resource library.

Therefore, it is necessary to use machine learning technology in college English teaching. In this context, this paper constructs an English teaching management data analysis system by introducing multi-source information and machine learning technology. By using the diversity of computers, we can improve the quality of online teaching, reduce the workload of teachers, and improve students' English language skills.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interests

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.

Data Availability

Data will be made available on request.

Qi GY (2016) The importance of English in primary school education in China: perceptions of students. Multiling Educ 6(1):1–18
Ting SH, Marzuki E, Chuah KM, Misieng J, Jerome C (2017) Employers’ views on the importance of English proficiency and communication skill for employability in Malaysia. Indonesian J Appl Linguistics 7(2):315–327
Bobanovic MK, Grzinic J (2011) The importance of English language skills in the tourism sector: A comparative study of students/employees perceptions in Croatia. Almatourism-Journal of Tourism Culture and Territorial Development 2(4):10–23
Khamidovna PO (2022) The role of interactive methods importance of english communication for engineering students. Spectr J Innov Reforms Dev 3:20–24
Liu J (2016) Study on the autonomous learning of college English based on online learning platform. Int J Smart Home 10(2):165–174
Mulyono H (2016) Using" Quipper" as an Online Platform for Teaching and Learning English as a Foreign Language. Teach Engl Technol 16(1):59–70
Sun Y, Jiang Y (1992) “Application of data mining in english online learning platform,” In Journal of Physics: Conference Series, Vol. No. 2, p. 022118, 2021
Es-Sabery F, Es-Sabery K, Hair A (2021) “A MapReduce improved ID3 decision tree for classifying twitter data,” In International Conference on Business Intelligence, pp. 160–182,
Li M (2018) A study on the influence of non-intelligence factors on college students’ English learning achievement based on C4. 5 algorithm of decision tree. Wireless Pers Commun 102(2):1213–1222
Yu J (2021) “Academic Performance Prediction Method of Online Education using Random Forest Algorithm and Artificial Intelligence Methods,”International Journal of Emerging Technologies in Learning, vol. 15, no. 5,
Prastikawati EF, Wiyaka W, Lestari MYW (2022) Secondary School Students’ Perception on Edmodo as Online Learning Platform in English Learning. Lang Circle: J Lang Literature 16(2):296–307
Zhang Y (2020) “Research on College English Online learning platform model based on big data technology,” In Journal of Physics: Conference Series, Vol. 1648, No. 4, p. 042090,
Onan A, Korukoğlu S, Bulut H (2016) A multiobjective weighted voting ensemble classifier based on differential evolution algorithm for text sentiment classification. Expert Syst Appl 62:1–16
Zhao Z, Jian Z, Gaba GS, Alroobaea R, Masud M, Rubaiee S (2021) An improved association rule mining algorithm for large data. J Intell Syst 30(1):750–762

Download PDF

Journal Publication

published 22 May, 2023

Read the published version in Soft Computing →

Reviewers agreed at journal
01 Apr, 2023
Editor assigned by journal
25 Mar, 2023
First submitted to journal
20 Mar, 2023

You are reading this latest preprint version

Application of English education big data system based on multi-source information fusion and machine learning

Status:

Journal Publication

Version 1

Abstract

Figures

1 Introduction

2 Related Work

3 Principle Of Multi-source Information Fusion And Machine Learning Recognition Algorithm

4 Architecture Design And Performance Analysis Of English Teaching Management Data Analysis System

4.1 System frame structure design

4.2 System function design

4.3 System data simulation performance test

5 Research On The Application And Development Strategy Of Education Big Data System

5.1 Existing problems of education big data application

5.2 Strategic design of education big data sharing platform

6 Conclusion

Declarations

References

Status:

Journal Publication

Version 1