Mobile QR Code QR CODE

  1. (School of Foreign Languages, Zhengzhou Tourism College, Zhengzhou 450000, China lihaifen@zztrc.edu.cn)



Visualization techniques, U-Net algorithm, 3D layout, Autonomous learning, Support rate

1. Introduction

Developing lifelong learning habits in students is one of the main responsibilities of education. However, due to the perennial implementation of exam-oriented education in China, teachers have become active players in teaching activities, while students have become passive learners, resulting in thousands of students not being able to form the ability to learn independently, let alone have the ability to learn independently. The current century is an era of information boom, and with the rapid development of information technology, the ways that learners access knowledge and information will change dramatically. Schools as well as teachers should encourage students to use information to carry out independent learning and to enhance their ability to use information technology to analyze and solve problems. In order to modernize education and meet students’ needs for independent learning, visualization has been created. Visualization technology has the ability to transform learned data and information into graphical information, which helps students increase their interest in learning. In addition, visualization technology can create a learning environment that facilitates visual perception, establishes a knowledge context, and promotes a deeper understanding of knowledge, thus improving students’ learning abilities. However, existing learning platforms are usually formal, and do not allow students to stay on the platform for long periods of time, making it difficult to learn in an efficient and immersive manner. Therefore, this study proposes an online, self-directed learning system based on visualization technology. At the same time, the study proposes a U-Net algorithm based on 3D layout extension optimization, considering that image enhancement helps to enhance the visual experience. The algorithm improves image feature information such as the contrast and fidelity in visual images from the teaching system so as to improve the quality of visual images and ultimately improve the effectiveness of the online learning platform.

2. Related Work

Many scholars have conducted a variety of studies on English language teaching at universities, including Thi Ngu D, who implemented English language teaching for history and economics courses, combining language teaching with images and ICT tools to obtain better subject matter and terminology. Sun et al. developed a deep-learning-assisted, online intelligent English teaching system by combining artificial intelligence modules with knowledge recommendations to help students improve the effectiveness of English teaching in line with knowledge acquisition and their own personality [6]. The results showed that the system was able to help students improve their learning efficiency while making the content more relevant. Xu proposed a strategy for building an English curriculum design based on personalized learning, hoping that students would become active learners in the classroom and develop harmoniously so that construction of the curriculum would be a virtuous ecological cycle [7]. Zhang and Perez-Paredes examined de-contextualization of postgraduate English learners’ use of mobile English learning resources [8]. Their study used a questionnaire and interviews to analyze 95 Chinese postgraduate students. The results showed that expanding their English vocabulary was an area that learners aimed to improve. C.~Li used a mixed methods approach to examine control value assessment as an antecedent of boredom in English learning among Chinese university students, and the results showed there is some complexity in the link between control value assessment and boredom [9]. Li has explored the fuzzy evaluation of college English teaching quality and established a fuzzy evaluation model of college English teaching quality based on analytic hierarchy process [10]. The implementation results show that the model can effectively assess the quality of English teaching in universities. H. Li developed an English learning system based on an improved fuzzy-assisted hierarchical neural network system, aiming to improve English education. The results showed that the system helped to improve the efficiency of classroom teaching [11]. Huiying and Qiang improved and analyzed the traditional MOOC based on cloud computing and artificial intelligence in order to improve efficiency by targeting the shortcomings of intercultural teaching of English in universities [12]. They improved the traditional algorithm according to the actual needs of MOOC teaching, and proposed a new improved model. Their research results showed that the model proposed had good performance and can effectively improve the efficiency of English intercultural teaching.

Many scholars have conducted research on 3D immersion techniques. Lan et al. tried to improve Chinese short essay writing skills of students who were learning Chinese as a second language in Singapore by using authentic contexts [13]. They conducted an experiment with the presence (or absence) of immersive second-language exploration in the prewriting phase of the activity as a control variable. Their results showed that students who explored authentic contexts prior to writing performed significantly better in constructing prewriting plans, and had significantly higher-quality writing than those who did not have such immersion experiences. The former group also showed higher motivation. Perin et al.~used immersive 3D imaging to make patients aware of their medical condition [14]. That study recruited 40 patients undergoing surgery for intracranial tumors, and showed that the experimental group using 3D, immersive informed consent had a better communication experience than the control group. Mulders et al. provided a framework for using immersive virtual reality in a learning environment based on the cognitive theory of multimedia learning [15], and showed that the method enhanced learners’ cognitive patterns. Mosher et al. provided social skills instruction based on augmented reality and mixed reality technology to school-aged students with autism spectrum disorders [16]. Their results produced recommendations from researchers and practitioners implementing AR and VR social skills interventions. Hanson et al. used immersive 3D technology to enhance understanding of complex science concepts, and showed that a 3D immersive learning experience was comparable to a 2D experience in terms of satisfaction and comfort, but it showed statistically significant improvements in post-test scores [17].

In summary, scholars have focused on the improvement of teaching quality and online intelligent English teaching systems for university English teaching. In terms of 3D immersion technology, scholars have focused on patient communication experiences and social skills instruction, but have conducted relatively little research on visualization in university English teaching. Therefore, this study builds an online, self-directed learning model based on visualization technology, image enhancement algorithms, and 3D immersion to enhance the learning experience.

3. Online Self-directed Learning Model Construction

3.1 Model Construction with Visualization

There are three basic assumptions that need to be set out in a visual model for online self-directed learning. First, the model should be constructed in such a way that learners are given the freedom to develop their own programs, rather than having a uniform curriculum set by teachers or colleges. This is because the latter do not differentiate learning programs to meet the needs of different types of students. Therefore, allowing learners to choose their own resources will increase flexibility and enthusiasm for learning. Secondly, the learning environment should uphold the principle of autonomy. Instead of the school fixing the learning environment for students, students should build the environment for their learning and learn independently in it. Third, to ensure the effectiveness of students’ self-learning while opening up free learning, rules and systems should be introduced to regulate activities in order to achieve a semi-structured teaching and learning design. Finally, the model should build a technological bridge for students to communicate with each other with the help of visualization technology in order to facilitate communication and discussion of students’ online learning. Fig. 1 shows a diagram of the online self-directed learning process.

Fig. 1. Sketch map of online autonomous learning teaching process.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig1.png
Fig. 2. Schematic diagram of five elements.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig2.png

The three-tiered structure of the online, self-directed learning process includes the topic planning tier, the program strategy tier, and the operational reflection tier. The three tiers are formed by a combination of five elements: learners, learning content, learning tools, learning activities, and learning objectives. The theme plan layer is responsible for defining the learning objectives and decomposing them; the solution strategy layer is responsible for combining the learning activities to achieve the learning objectives set out in the theme plan layer. The operational reflection layer is responsible for combining the three elements of learning content, learning tools, and learners to form a complete learning activity. The three-tier structure presented in Fig. 1 can be seen as progressively refined and is broken down from left to right, thus providing an effective guide to student learning behavior. The five elements also create a differentiated learning environment because of the multiple ways in which they converge. Fig. 2 illustrates the relationships among the five elements.

Learners denote the group of students involved in a learning activity. Because online independent learning emphasizes the active nature of learning, students are no longer just passive recipients but managers and creators of learning activities as well as commentators and consumers of teaching and learning practices. The reasonableness of learning objectives is directly related to the outcomes that learners ultimately achieve. The reasonableness of learning objectives is reflected in their operability and consistency with learners’ internal values. From the perspective of content, learning objectives can be categorized as knowledge-based, skills-based, and affective; from the perspective of difficulty, learning objectives can be categorized as growth-based and maintenance-based. Learning activities can only be carried out if objectives are well formulated. Online independent learning greatly enriches the range of activities, enabling learners to personalize and differentiate their learning. For example, activities can be divided into research, collaborative, and thematic learning. Online self-directed learning broadens the concept of learning content. As a result, learning content is no longer confined to the strictly organized and logical scope of traditional teaching and learning, but extends to the vast resources available throughout the Internet. Learning tools are responsible for helping learners carry out their activities, and are divided into two categories: specialized tools and generic tools. Fig. 3 shows a schematic of the technology used to visualize the online independent learning process.

Fig. 3. Schematic of online autonomous learning visualization technology.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig3.png

Fig. 3 shows that visualization of the online, self-directed learning process is structured in four layers; from top to bottom: the data input layer, the visual coding layer, the logical recommendation layer, and the view presentation layer. The data input layer includes role data, such as learners, teachers, and teaching assistants; learning content data, such as forum statements and student assignments; and work data, such as learning progress and learning contexts. These three types of data are communicated to form the learning data, and are used to reflect the current state of learning. The visualization and coding layer is responsible for analyzing the data, using statistical techniques, text visualization techniques, or human-computer interaction techniques to code the input data in order to visualize them. The logical recommendation layer provides students with references to learning solutions based on learner characteristics, learning tasks, and knowledge networks, helping them to build a personalized learning environment. The view presentation layer is related to visualization of the content space, and is used to present information such as views of the learning network or a list of learning activities.

3.2 Optimization of Immersive Networks with the U-Net Algorithm and a 3D Layout

Due to the overwhelming size of the teaching data, this research first designed a visual enhancement algorithm based on the U-Net neural network in order to enhance visualization of the network data and improve the user’s visual and other sensory experiences. Then, based on this algorithm, this study proposes an immersive, network topology visualization layout technique reflecting a 3D immersion effect. Fig. 4 shows a schematic of the U-Net neural network structure.

Fig. 4. The U-Net neural network structure.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig4.png

As seen in Fig. 4, the network consists of an encoder unit, a decoder unit, and a hop-level connection unit. The encoder and decoder units are symmetrical. Both are formed by a convolution module. The function of the convolution module is to capture the deeper features of the visual image. The role of the hopping connection unit is to enhance details of the image based on deep feature information in the visual image. Eq. (1) is the mathematical expression used for smoothing and denoising in this network.

(1)
$ \begin{array}{l} y'=med\left\{I\left(\alpha _{1},\beta _{1}\right),I\left(\alpha _{2},\beta _{2}\right),\ldots ,I\left(\alpha _{g},\beta _{g}\right)\right\}\\ =\left\{\begin{array}{l} I\left(\alpha _{\left(\frac{g+1}{2}\right)},\beta _{\left(\frac{g+1}{2}\right)}\right),g=2a+1\\ \frac{1}{2}*\left[I\left(\alpha _{\left(\frac{g}{2}\right)}+\beta _{\left(\frac{g}{2}\right)}\right)\right],g=2a \end{array}\right. \end{array} $

In Eq. (1), $y'$ is the median grey value of an image pixel point; $med$denotes the median calculation function; $I(\alpha _{g},\beta _{g})$denotes a pixel point; $a$ denotes an arbitrary integer; $g$ denotes the number of rows of pixel points; and $(\alpha _{g},\beta _{g})$, $\alpha _{g}$, and$\beta _{g}$ represent the position coordinates, abscissa, and ordinate of the pixel points, respectively.

Eq. (2) is the mathematical expression for image edge sharpening:

(2)
$ grad\left(t\right)=\frac{\partial t}{\partial \alpha }a+\frac{\partial t}{\partial \beta }c+\frac{\partial t}{\partial \gamma }e $

In Eq. (2), $a$, $c$, and $e$ all represent arbitrary constants; and $grad$ denotes the vector field gradient. Eq. (3) represents the expression for the gradient corresponding to the vectorially defined coordinates:

(3)
$ grad\left[I\left(\alpha ,\beta \right)\right]=\left[\begin{array}{l} \frac{\partial I}{\partial \alpha }\\ \frac{\partial I}{\partial \beta } \end{array}\right] $

In Eq. (3), $grad\left[I(\alpha ,\beta )\right]$ is the gradient value corresponding to pixel $I(\alpha ,\beta )$. In terms of dimensionality, the network topology can be divided into three types: a one-dimensional layout, a two-dimensional layout, and a three-dimensional layout. One-dimensional layouts have the advantage of establishing the overall relevance of the network, but weaken the relationships between information within the community. Given the increasing complexity of the network structure and its strong, time-varying nature, two-dimensional layouts also make it difficult to meet practical needs. Therefore, this study uses an edge aggregation strategy to extend the two-dimensional layout to a three-dimensional layout.

Suppose the number of network nodes is $n$, the number of network edges is$m$, and the degree corresponding to any node $v$ is $D_{v}$. The network can be considered undirected graph adjacency matrix $A$, whereupon the modular degree can be expressed by Eq. (4):

(4)
$ Q=\frac{1}{2m}\sum _{ij}\left[A_{ij}-\frac{D_{i}D_{j}}{2m}\right]\delta _{ij} $

In Eq. (4), $Q$ indicates the module degree, and $\delta _{ij}$ indicates whether the nodes belong to the same community. The purpose of using the module degree is to divide the network, and when the module degree iterates to the maximum value, it indicates the end of network division. In order to avoid confusion in the layout of network nodes and to avoid a crossover problem for network edges, this study introduces an energy model to simulate the network layout process once the network communities are determined. Eq. (5) is the mathematical expression of the energy model:

(5)
$ E=\sum _{i=1}^{n}\sum _{j=1}^{n}\frac{1}{2}k\left[d\left(i,j\right)-s\left(i,j\right)\right]^{2}+\sum _{i=1}^{n}\sum _{j=1}^{n}\frac{rw_{ij}}{d\left(i,j\right)^{2}} $

In Eq. (5), $d(i,j)$ represents node spacing; $s(i,j)$ represents the spring model length; $k$ and $r$ represent the elasticity coefficient and electrostatic force constant, respectively; and $w$ denotes the inter-node weights. Let the set of supernodes for the 3D network be $C$\textcolor{color-6}{.} The coordinates of the supernodes, $C_{i}$, after the energy model layout are $(x_{i},y_{i},z_{i})$. In order to reduce the loss of community structure features, the network needs to be laid out twice with the supernodes as coordinate centers. Then, the spatial translation change matrix at this point is Eq. (6):

(6)
$ T\left(t\right)=T\left(t_{x},t_{y},t_{z}\right)=\left[\begin{array}{l} 100t_{x}\\ 010t_{y}\\ 001t_{z}\\ 0001 \end{array}\right] $

In Eq. (6), $(t_{x},t_{y},t_{z})$ is the spatial translation vector. After quadratic layout, the nodes’ flush coordinate matrix in 3D space can be obtained with Eq. (7):

(7)
$ P\left(*\right)=\left\{\begin{array}{l} T\left(t\right)\times \left[x_{i} y_{i} z1\right]^{T},if1\text{D}\\ T\left(t\right)\times \left[x_{i} y z1\right]^{T},if2\text{D}=\left[\begin{array}{l} x*\\ y*\\ z*\\ 1 \end{array}\right]\\ T\left(t\right)\times \left[x y z1\right]^{T},if3\text{D} \end{array}\right. $

In Eq. (7), $T$ represents the spatial translation change matrix; $\left[x_{i} y_{i} z1\right]$ denotes a one-dimensional chi-square coordinate matrix; $\left[x_{i} y z1\right]^{T}$denotes a two-dimensional chi-square coordinate matrix; and $\left[x y z1\right]^{T}$denotes a three-dimensional chi-square coordinate matrix. Afterwards, the study uses an edge aggregation strategy to implement an immersive 3D edge-bound layout. The strategy consists of four basic rules, shown as a schematic diagram in Fig. 5.

Fig. 5. Schematic of the four basic rules in the edge aggregation strategy.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig5.png

Rule 1 is a directionally compatible strategy with the mathematical expression in Eq. (8):

(8)
$C_{a}(P,Q)=\left| \cos (\alpha )\right| ,where \alpha =\arccos (\frac{P\cdot Q}{\left| P\right| \left| Q\right| })$

In Eq. (8),$C_{a}(P,Q)$ denotes the directional compatibility size, $\alpha $ represents the angle between two edges, while $P$ and $Q$ represent the two edges. If the directional compatibility size is 1, the two edges are perpendicular, and if the directional compatibility size is 0, the two edges are parallel. Rule 2 is the size compatibility strategy; the mathematical expression is Eq. (9):

(9)
$ C_{s}\left(P,Q\right)=\frac{2}{l_{avg}\cdot \min \left(\left| P\right| ,\left| Q\right| \right)+\max \left(\left| P\right| ,\left| Q\right| \right)/l_{avg}} $

In Eq. (9), $C_{s}(P,Q)$ indicates the size of the dimensional compatibility, and $l_{avg}$ is the average length of the two edges. If the compatibility size is 1, the two edges are of equal length; if the compatibility size is 0, the two edges are in a size-incompatible relationship. Rule 3 is a positional compatibility policy with its mathematical expression in Eq. (10):

(10)
$ C_{p}\left(P,Q\right)=\frac{l_{avg}}{l_{avg}+\left\| P_{m}-Q_{m}\right\| } $

In Eq. (10), $P_{m}$ and $Q_{m}$ represent the midpoints of the two edges, $P$ and $Q$. When the position compatibility size is 1, the two edges intersect; when the position compatibility size is 0, the two edges never intersect. Rule 4 represents the visual compatibility policy for which the mathematical expression is Eq. (11):

(11)
$ C_{v}\left(P,Q\right)=\min \left(V\left(P,Q\right),V\left(Q,P\right)\right) $

in which $V(P,Q)=\max (1-2\left\| P_{m}-I_{m}\right\| \div \left\| I_{0}-I_{1}\right\| ,0)$ where $I_{m}$ is the midpoint of $I_{0}$ and $I_{1}$.

Eq. (12) is the overall compatibility calculation formula:

(12)
$ C_{e}\left(P,Q\right)=C_{a}\times C_{s}\times C_{p}\times C_{v} $

Eq. (12) combines the above four compatibility edge binding rules to obtain overall compatibility $C_{e}(P,Q)$. At this point, $C_{e}(P,Q)\in [0,1]$.

4. Utility of the Online Independent Learning Visual Teaching System

4.1 Analysis of Algorithm Performance

In order to verify the effectiveness of the U-Net algorithm based on 3D layout extension optimization for visual image enhancement, two datasets were chosen: the LOL dataset and the LIME dataset. For quantitative evaluation of the visual image enhancement effect, peak signal-to-noise ratio (PSNR), structural similarity (SSIM), image entropy (IE) and contrast gain (CG) were chosen as the performance metrics. PSNR measures the level of image distortion; SSIM measures the similarity between the reference image and the enhanced image in terms of luminance, contrast, and structure; IE measures the richness of information contained in the image; and CG describes the magnitude of the contrast gain after image enhancement. Fig. 6 shows the results of the PSNR and SSIM metrics for different algorithms tested on the LOL dataset.

From Fig. 6(a), we can see that the proposed algorithm achieved higher PSNR values for the different images tested, which indicates less distortion. For the desk image, the PSNR of the proposed algorithm was 22.24 while PSNR of the MBLLEN, RetinexNet, and Dong algorithms were 22.13, 19.77, and 15.90, respectively. In addition, for the mean PSNR of multiple images, the proposed algorithm obtained the highest value (21.67) while the mean PSNR for MBLLEN, RetinexNet, and Dong were 21.23, 20.19, and 18.53, respectively. Fig. 6(b) shows that the algorithm from this study obtained the highest SSIM for the different images tested, indicating that the images enhanced by this algorithm are more similar to the original images. Mean SSIM from the proposed algorithm was 0.80, while SSIM obtained by the MBLLEN, RetinexNet, and Dong algorithms were 0.79, 0.69, and 0.63, respectively. Thus, improvements by the proposed algorithm were 0.01, 0.11, and 0.17 over the other three. IE and CG metrics showed similar results.

Fig. 6. PSNR and SSIM results from the algorithms tested against the LOL dataset.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig6.png
Fig. 7. IE and CG results from the algorithms tested against the LIME dataset.
../../Resources/ieie/IEIESPC.2024.13.5.490/fig7.png

From Fig. 7(a), we can see that the U-Net algorithm based on 3D layout expansion optimization proposed in this study obtained higher IE values in tests with different images, which indicates that the proposed algorithm is more capable of enhancing the information richness of the images. However, for the same test on different images, the MBLLEN, RetinexNet, and Dong algorithms obtained IE averages of 7.62, 7.35, and 7.13, respectively, which the proposed algorithm improved by 0.05, 0.32, and 0.54, respectively. Fig. 7(b) shows the proposed image enhancement algorithm obtained higher CG values, which indicates the algorithm is more capable of enhancing the contrast of an image. The mean CG obtained by the proposed algorithm was 3.95, while mean CG obtained by the MBLLEN, RetinexNet, and Dong algorithms were 2.67, 2.43, and 2.12, respectively. So, the proposed algorithm improved the CG metric by 1.28, 1.52, and 1.83, respectively.

4.2 Analysis of the Degree of Support from the Visual Teaching System

To find out how students evaluated the visual teaching techniques proposed by this study, a questionnaire investigated the most preferred visual views. The results are shown in Table 1.

Table 1. The most popular visual views.

Serial No.

Visual form

Number of votes

1

Experience ranking

59

2

Rankings for learning task completion

53

3

Empirical radar

46

4

Final estimated score

39

5

Knowledge map

35

6

Q & A discussion interaction diagram

32

As can be seen from Table 1, the top three visualization views preferred by students were the experience value ranking chart, the learning task completion ranking chart, and the experience value radar chart, which received 59, 53, and 46 votes, respectively. It is clear from this that students are more interested in learning with progress-related views.

In order to verify the validity of the study’s proposed online self-directed learning model, the Self-Directed Learning Support Scale tested the university students’ support for the model in four dimensions of self-directed learning (cognitive, motivational, affective, and social), and in three dimensions of the learning process (prior planning, monitoring behavior, and evaluating reflection). Table 2 shows those results.

Table 2. Support results.

Dimension

Number of people

Minimum

Maximum

Mean value

Standard deviation

Cognition

64

2.01

5.00

4.13

0.756

Motivation

64

1.74

5.00

3.80

0.843

Emotion

64

1.86

5.00

3.93

0.872

Social contact

64

1.94

5.00

4.05

0.781

Planning ahead

64

1.92

5.00

3.97

0.807

Monitoring behavior

64

1.89

5.00

3.83

0.918

Evaluation reflection

64

1.75

5.00

3.79

0.849

Table 2 shows that the mean support ratings for the four dimensions of independent learning were 4.13, 3.80, 3.93, and 4.05, respectively, while the mean support ratings for the three dimensions of the learning process were 3.97, 3.83, and 3.79, respectively. We can see that the support ratings given for the model were all higher than 3.7, which indicates that students rated the online independent learning model as quite good. The cognitive dimension received the highest support rating from students for independent learning, with a score of 4.13, and the realization plan dimension received the highest support rating for the learning process with a score of 3.97. Table 3 shows statistics for the effectiveness of the visualization technology in supporting the learning process.

Table 3. Support effect from visualization technology on the learning process.

Evaluation dimension

Evaluation of project

Fully supported (%)

Supported (%)

Commonly supported (%)

Not Supported (%)

Not supported at all (%)

Learning adaptability

Helps to quickly adapt to the learning system

51.6

48.4

0

0

0

Helps to plan learning time reasonably

59.4

37.5

1.6

1.6

0

Helps learning discovery and exploration

46.9

48.4

0

0

4.7

Helps promote learning

40.6

51.6

6.2

0

1.6

Learning attention

Helps me keep learning

39.1

61.9

0

0

0

Increases learning efficiency

56.3

35.9

3.1

4.7

0

Helps with learning transfer

51.6

39.1

7.58

1.6

0

Helps focus on learning

67.2

29.7

1.6

1.6

0

As can be seen in Table 3, in terms of visualization technology improving students’ adaptation to learning, the most recognized aspect is that visualization technology helps to plan study time rationally, with a Fully Supported rating of 59.4%. At the same time, visualization technology supported all students by helping them adapt to the learning system quickly, with a Fully Supported rating of 51.6% and a Supported rating of 48.4%. In terms of visualization technology helping students to improve their concentration, the most recognized was that visualization technology helped to focus learning, with a Fully Supported rating of 67.2%. Therefore, the visualization techniques used in this study are effective in enhancing learning adaptation as well as learning concentration.

5. Conclusion

The development of information technology has transformed teaching and learning activities, with online self-directed learning becoming a new trend. Visualization technology has the advantage of enhancing the visual perception of learners, helping them to construct their own learning environments and learning tasks. Therefore, this study constructed an online, self-directed learning model based on visualization technology, and optimized visualization by using the U-Net algorithm based on a 3D layout extension. The results showed that the visual image enhancement algorithm proposed in this study achieved higher values in both PSNR and SSIM when tested on the LOL dataset, indicating that visual image distortion by the algorithm is relatively small. In the student support test of the system, students enjoyed high support from the system’s assistance in autonomous learning and the learning processes, with scores above 3.7. In terms of improving students’ learning adaptability through visualization technology, students gave a full-support rating of 59.4% for the help provided by visualization technology in reasonable planning of learning time. This indicates that the visual teaching model proposed by the research institute helps to improve students’ self-efficacy, stimulates learning motivation, and increases learning strategies to improve online autonomous learning behavior.

One deficiency in this study is that it fails to consider compatibility and integration of cross-platform learning resource forms. In addition, there is room for improvement in data granularity of the learning behavior analysis. Therefore, in future work, more development framework technologies and deep learning algorithms can be used to achieve multi-stage optimization.

REFERENCES

1 
Yin Y M, Mu G M. Examination-oriented or quality-oriented? A question for fellows of an alternative teacher preparation program in China. The Australian Educational Researcher, 2022, 49(4): 727-742.DOI
2 
Ariebowo T. Autonomous learning during COVID-19 pandemic: Students’ objectives and preferences. Journal of Foreign Language Teaching and Learning, 2021, 6(1): 56-77.DOI
3 
Chenwu X U, Yang Z H. Research on the Autonomous Learning Mode of College Students Based on Mobile Terminals under the Internet Environment. Transactions, 2021, 3: 23-26.URL
4 
Sun S. Construction of integrated cloud computing platform for Japanese autonomous learning and examination. computer Applications in Engineering Education, 2021, 29(2): 312-320.DOI
5 
Thi Ngu D, Huong D T, Huy D T N, Thanh P T, Dongul E S. Language teaching application to English students at master’s grade levels on history and macroeconomic-banking management courses in universities and colleges. Journal of Language and Linguistic Studies, 2021, 17(3): 1457-1468.URL
6 
Sun Z, Anbarasan M, Praveen Kumar D. Design of online intelligent English teaching platform based on artificial intelligence techniques. Computational Intelligence, 2021, 37(3): 1166-1180.DOI
7 
Xu Y. A study on the curriculum design of personalized English teaching in the context of Internet. Open Journal of Social Sciences, 2021, 9(4): 65-71.DOI
8 
Zhang D, Perez-Paredes P. Chinese postgraduate EFL learners’ self-directed use of mobile English learning resources. computer Assisted Language Learning, 2021, 34(8): 1128-1153.DOI
9 
Li C. A Control-Value Theory approach to boredom in English classes among university students in China. The Modern Language Journal, 2021, 105(1): 317-334.DOI
10 
Li N. A fuzzy evaluation model of college English teaching quality based on analytic hierarchy process. International Journal of Emerging Technologies in Learning (iJET), 2021, 16(2): 17-30.URL
11 
Li H. Improved fuzzy-assisted hierarchical neural network system for design of computer-aided English teaching Computational Intelligence, 2021, 37(3): 1199-1216.DOI
12 
Huiying X, Qiang M. College English cross-cultural teaching based on cloud computing MOOC platform and artificial intelligence. journal of Intelligent & Fuzzy Systems, 2021, 40(4): 7335-7345.DOI
13 
Lan Y J, Lyu B N, Chin C K. Does a 3D immersive experience enhance Mandarin writing by CSL students? Language Learning & Technology, 2019, 23(2): 125-144.DOI
14 
Perin A, Galbiati T F, Ayadi R, Gambatesa E, Orena E F. Informed consent through 3D virtual reality: a randomized clinical trial. acta neurochirurgica, 2021, 163(2): 301-308.DOI
15 
Mulders M, Buchner J, Kerres M. A framework for the use of immersive virtual reality in learning environments. international Journal of Emerging Technologies in Learning (IJET), 2020, 15(24): 208-224.URL
16 
Mosher M A, Carreon A C, Craig S L, Ruhter L C. Immersive technology to teach social skills to students with autism spectrum disorder: A literature review. Review Journal of Autism and Developmental Disorders, 2022, 9(3): 334-350.DOI
17 
Hanson J, Andersen P, Dunn P K. Effectiveness of three-dimensional visualization on undergraduate nursing and midwifery students’ knowledge and achievement in pharmacology: a mixed methods study. Nurse Education Today, 2019, 81: 19-25.DOI
Haifen Li
../../Resources/ieie/IEIESPC.2024.13.5.490/au1.png

Haifen Li was born in Puyang, Henan, PR China, in 1982. She received the master's degree from Zhengzhou University, PR China. Now, she works in School of Foreign Languages Department, Zhengzhou Tourism College. Her main research area is English teaching and translation.