Using Information Technology to Optimize the Allocation of Teaching Resources in Open Education and Educational Reform Practices

In recent years, with the rapid development of information technology and the popularization of the Internet, open educational resources have become a new favorite in the field of education. Open educational resources refer to free, open, and freely available educational resources provided to students and teachers through the Internet and other channels. They provide more learning resources and communication platforms for students and teachers, but at the same time, they also bring some challenges [1–4].

First of all, OER breaks the barriers of traditional educational resources and enables students and teachers to access and share knowledge all the time. Students can independently choose the content and learning methods according to their own interests and needs [5–7]. Teachers can also find suitable teaching materials and methods from the rich and diverse resources. In addition, OER also promotes the sharing of resources and cooperation in global education. On the one hand, it promotes global educational equity through sharing of educational resources. On the other hand, it improves the quality of education through cooperation in teaching and research across national boundaries [8–11]. However, the utilization of OER also faces some challenges. The first one is the copyright issue. The openness of OER makes it easier to use and modify the resources, but it is also prone to copyright infringement and piracy problems [12–14]. This poses a certain threat to the rights and interests of creators. The second is the assurance of resource quality. Due to the massive generation and wide circulation of OER, the quality of the resources cannot be fully guaranteed, and the problems of troubled choices and insufficient funding are also important factors that hinder the utilization of students and the sustainable development of OER [15–18].

The research optimizes the allocation of teaching resources in open education in two aspects. The open education teaching resources sharing method based on blockchain technology constructs a resource sharing model from three levels: application layer, contract layer, and data layer. The personalized recommendation method of open education teaching resources based on K-Means clustering, firstly, divides the personalized recommendation level of open education resources according to the common preferences between two groups of adjacent users. Second, the collaborative filtering algorithm is selected to normalize the sample data, calculate the similarity to formulate user preferences, and construct the open education resources recommendation model. Finally, based on the K-Means clustering algorithm given the objective function, the personalized recommendation process is established to achieve the recommendation of open educational resources.

2

Methodology for optimizing the allocation of teaching resources in open education

2.1

Characteristics of open education teaching resources

2.1.1

No temporal or spatial constraints

Open education teaching has been transformed from the traditional teacher-teaching-based to learner-directed learning-based, so that the learner’s scope is wider, and can be free from both time and space constraints, to independently determine the pace of learning, independently select the learning content, the reasonable choice of learning tools, and carry out personalized learning [19].

2.1.2

Coexistence of multiple types of teaching resources

Open education attaches more importance to the teaching process, and teaching resources are more diverse, such as online courses, thematic learning websites, microclasses, and so on. Focusing on the teaching of microcourses, based on its content characteristics, presentation mode, target audience, learning strategies, etc., we effectively integrate a variety of learning resources to build a microcourse teaching resource package with reasonable structure, clear content and precise positioning, which can assist learners to carry out effective learning.

2.1.3

Emphasize learner initiative in the use of resources

Learners’ own learning needs, personality characteristics, and so on will have a great impact on their learning motivation and learning effect. Open education teaching resources in the context of the “Internet Plus” era emphasize learners’ initiative in using the resources. Learners can not only learn according to their own needs, but also personalized learning according to their own personality characteristics, and can strengthen their learning by checking and making up for deficiencies, which is conducive to the realization of lifelong learning.

2.2

Methods of optimizing the allocation of teaching resources

According to the previous analysis of the characteristics of open education teaching resources, this chapter proposes the establishment of a resource sharing mechanism and personalized recommendation of resources to achieve the optimization of teaching resource allocation.

2.2.1

Optimization of Teaching Resource Allocation Based on Sharing Models

1)

Integration of open education teaching resources

In order to ensure that open education teaching resources can play a greater role in the actual application, this paper integrates the existing open education teaching resources, and the integration function expression is as follows: 1 $K = \sum_{i > 1}^{n} k_{i} \times L_{i}$ where K is the result of open education teaching resources integration. k_i is the integration interface for the OER. L_i is the length of the OER. i is the category of the OER.

2)

Construction of open education teaching resource sharing model based on blockchain technology

This paper utilizes blockchain technology to construct an open education teaching resource sharing model. Blockchain technology, a distributed database technology, enables the storage and transmission of highly secure data through decentralization and tampering. The application of blockchain technology in teaching resource sharing model building can prevent data from being tampered with or destroyed. Based on blockchain technology, this paper divides the resource sharing model into three parts: the application layer, the contract layer, and the data layer. Among them, there are multiple functions in the application layer to enable the uploading, downloading, and viewing of teaching resources. The contract layer contains multiple smart contracts, which can ensure the integrity and security of teaching resources [20]. The data layer is able to utilize the coalition chain to deploy the above contracts onto the blockchain, thereby facilitating the sharing of teaching resources. In order to ensure the sharing performance of the sharing model, the transmission rate of teaching resources must be ensured, and the transmission efficiency and transmission channel of teaching resources are determined, and the calculation process is as follows: 2 ${\begin{array}{l} Z = \sum_{c > 1}^{M} \log (1 + α_{c}) + u_{M} \\ L = κ_{i} \sum_{i > 1}^{n} λ_{i} \times u_{i} \end{array}$ where Z is the rate of teaching resource transmission. α_c is the transmission factor of teaching resources. u_M is the maximum number of times a teaching resource can be captured. L is the communication channel for the teaching resources in the shared model. κ_i is the format in which the teaching resources are transmitted in this communication channel. λ_i is the usage of the communication channel. u_i is the communication channel network traffic.

3)

Sharing of Open Education Teaching Resources

The sharing of open education teaching resources is realized by using the open education teaching resources sharing model constructed above. The integration process of open education teaching resources is as follows: 3 $F_{i} = \frac{1}{x^{2}} (1 - K) D_{i} H_{i}$

where F_i is the integration result of different categories of open education teaching resources. x is the transmission link of teaching resources, D_i is the integration instruction of computer, and H is the transmission bandwidth of teaching resources. Through the above formula, this paper completes the integration processing of open education teaching resources, and realizes the sharing of teaching resources according to the integration results, combined with the sharing model constructed above. The sharing process is as follows: 4 $y_{d} = \frac{P_{d} G_{d}}{N_{0} + \sum Y_{d} P G_{d c}} \times F_{i}$

Where y_d is the sharing result of teaching resources. P_d is the uncontrollable interference in the transmission process of teaching resources. G_d is the shared instruction of the teaching resource transmission process. N₀ is the initial amount of shared resources. Y_d is the sharing gain value of instructional resources. P is the bandwidth of the instructional resource sharing channel. G_dc is the sharing protocol of teaching resources.

2.2.2

Optimization of teaching resource allocation based on personalized recommendation

1)

Classifying the personalized recommendation level of open education resources

The realization of personalized recommendations in open education can provide more learning materials to target users, and also provide users with the possibility of independent learning. In the face of a large number of open materials in the database, users are unable to choose a suitable course according to their own situation at the first time, which leads to blindness in open education. In order to realize the personalized recommendation of open education resources, this time, we choose the user’s preference extraction technology to classify the personalized recommendation level of open education resources, and on this basis, we choose the collaborative filtering algorithm to construct the recommendation model.

2)

Collaborative filtering algorithm to construct the resource recommendation model

The process of user preference extraction can take the behavioral habits of specific users as a support point, take the preference of students with higher learning level as a recommendation type, follow up to other target users, and analyze the user’s own characteristics under the personalized characteristics of college students to find the adjacent attributes between different users [21].

The collaborative filtering algorithm is chosen to establish the open education resources recommendation model, and the steps are divided into data behavior collection and similarity calculation. Among them, it is necessary to ensure the diversity of data in the collection of user behavior data, collect user preference data in various forms, constrain the data collection within [0,1], and carry out normalization to achieve the similarity calculation, which is calculated by the formula: 5 $a (s, d) = \sqrt{\sum {(s_{f} - d_{f})}^{2}}$ 6 $s i m (s, d) = \frac{1}{a (s, d)}$ where: s, d are any two points in dimensional space, respectively. f is the number of points. a(s, d) is the Euclidean distance between s and d. The distance is used to calculate the similarity quantitative sim(s, d), so that the distance between s and d can directly represent the user’s quantitative preference for a certain kind of open educational resources, and the K-Means clustering algorithm is selected to achieve personalized recommendation of resources according to the user’s quantitative preference.

3)

Recommending Open Educational Resources Based on K-Means Clustering

Personalized recommendation is realized by K-Means clustering method, g user data are randomly selected as the initial center, and the distance of other objects to the initial clustering center of g group is calculated. Where l is the sample object, set the objective function as: 7 $h_{j} = \sum_{k = 1}^{g} \sum_{l \in j_{k}} {| l - c_{k} |}^{2}$

Where: j_k is the subclass of the data object. c_k is the mean value of j_k. h_j is a function of l and c_k and can be expressed as the squared difference between the two. In the case of l determination, c_k can vary with l, as shown by the fact that the value of h_j is positively correlated with the clustering error and inversely correlated with the clustering result. After losing user objects and clustering centers, multiple clustering centers can be randomly selected according to the clustering algorithm to classify the smallest data objects into that type, and when the clustering centers are exactly the same, the recommendation of open educational resources can be completed.

3

Experimental analysis of the allocation of teaching resources

3.1

Analysis of Resource Sharing Experiments

In order to prove the practicality of the open education teaching resource sharing model based on blockchain technology, the following experiments are designed.

The model is used to conduct sharing experiments on distributed teaching resources in multi-campus in a region, and the experimental results of resource uploading efficiency are shown in Figure 1. As the total amount of teaching resources increases, the speed of resource uploading gradually decreases after using the model in this paper, but its decline is not large, and the difference between the highest and lowest values of resource uploading efficiency is only 8%, which indicates that its sharing speed is fast.

Assuming that there are 1000 resources that need to be updated in the multi-campus distributed teaching resources, the original teaching resources are set to be divided into 6 groups, and at the same time, there exists a different upper limit of updating in each group. Using the model of this paper to share teaching resources, the results of sharing teaching resources are obtained as shown in Fig. 2. After using the model of this paper to share the teaching resources, the difference of each group from 1 to 6 groups is 4, 3, 2, 5, 1, 1, respectively, and the updating amount of the teaching resources is not much different from the updating upper limit, which indicates that it is updated quickly.

On this basis, user feedback on the model of this paper is explored. The study was carried out in the form of a questionnaire, as shown in Table 1. Teachers and students gave high ratings to the sharing model’s resource real-time update effect, resource richness, access speed, and resource search accuracy. The model’s on-demand recommendation effect requires improvement.

Table 1.

The user feedback situation of the model statistics

Evaluation dimension	Teacher feedback	Student feedback
Resource real-time update effect	Superior	Superior
Resource richness	Superior	Superior
Resource access speed	Superior+	Superior+
Resource search accuracy	Superior+	Superior+
The resources are recommended for recommendation	General	General

3.2

Experimental Analysis of Personalized Recommendation

In order to verify the effectiveness of the article’s proposed K-Means clustering-based personalized recommendation method for open education teaching resources in practical applications, users adopting open education teaching resources were selected as the research subjects, and a comparison test was conducted. Two comparison methods were chosen for this test, namely, the personalized recommendation method of teaching resources based on cognitive diagnosis (Method 1) and the personalized recommendation system of teaching resources based on data mining (Method 2). In the comparison test, personalized recommendation accuracy and recommendation time were taken as the test indexes, and the method proposed in the article was used to test with two comparison methods.

3.2.1

Experimental indicators

Accuracy is a measure of how many of the recommendations given by a recommendation method are teaching resources that the user is really interested in, indicating the proportion of all teaching resources recommended to the user that the user is really interested in.

The time required for the recommendation method to react to complete the recommendation process after the user submits a request is known as recommendation time. A shorter recommendation time improves the user experience and enables users to access the teaching resources they need as soon as possible.

3.2.2

Analysis of experimental results

Using three methods for personalized recommendation of open education teaching resources and comparing the recommendation accuracy, the results are obtained as shown in Fig. 3. With the increase in the number of experiments, the recommendation accuracy of this paper’s method continues to improve. After a number of experiments, the recommendation accuracy of this paper’s method is higher than 93.6%, and it is relatively smooth, and the highest can even reach 96.2%. The recommendation accuracy of the two comparative methods is less than 84.2% by contrast. This paper’s method is proven to be more effective in providing personalized recommendations for open education teaching resources.

Then, the test is carried out with personalized recommendation time as an index, and the comparison results of the three methods are shown in Figure 4. The personalized recommendation time of applying the method of this paper is the highest 21.8s, while the personalized recommendation time of the two methods in comparison is the highest 43.9s, which proves that the method of this paper is more efficient in recommending teaching resources.

4

Analysis of the effects of educational reform practices

Applying the two models designed in this paper to open education teaching resource allocation. Take open education learners who are undergoing English writing enhancement as the research subjects. The experiment is divided into a control group and an experimental group, both of which have 32 participants. The control group adopts the traditional way of learning before teaching reform, and the experimental group adopts the way after resource allocation optimization.

4.1

Experimental design

This study used the data analysis and statistical software SPSS17.0 Chinese version to correlate the two raters’ pre- and post-test compositions and to test the consistency of their marking criteria. The essays were evaluated in four dimensions: content, structure, grammar, and vocabulary (including word usage, word spelling, punctuation, and capitalization). Each part of the evaluation received 5 points out of a total of 20 points. The average score given by the two teachers was the final score of the study participants. However, if the difference between the scores given by the two teachers is more than 4 points, the third teacher in the subject group will reevaluate the essay and then take the average of the two closest scores as the final score of the subject. In order to verify the consistency of their marking standards, the researcher analyzed the correlation between the scores given by the two raters, and the results were as follows: the Pearson correlation coefficient of the total scores of the assignments given by the two raters was 0.733, which reached the level of significant correlation, indicating that the two raters had uniform standards and high consistency in their marking. Independent samples t-tests were conducted on the total pre-test and post-test scores and the dimension scores of the pre-test and post-test between the two groups in the experimental group and the control group to test whether there was a significant difference in the total pre-test and post-test scores and content scores between the two groups. Paired samples t-tests were conducted on the total scores and dimension scores of the pre- and post-tests within the experimental group and the control group, respectively, to test whether there were significant differences in the total scores and content scores of the pre- and post-tests within the experimental group and the control group. After 4 months, a questionnaire was administered to the students in the experimental group to find out their views on the relationship between the optimal allocation of teaching resources in open education and the improvement of writing skills.

4.2

Experimental results

4.2.1

Independent samples t-test results

The results of the between-group comparison are shown in Table 2. The difference in total pretest score between the experimental and control groups was 0.03 points, and the difference in pretest content, structure, grammar, and vocabulary scores was 0.21, 0,06, and 0.23 points respectively. Independent samples t-tests were conducted on the pretest data of the two groups respectively, and the results proved that the p-values of the total pretest scores, content, structure, grammar, and vocabulary scores were all greater than 0.05, which indicated that the differences existing between the two groups in the total pretest scores and scores of the dimensions were not statistically significant. That is, the students in the two groups, the experimental group and the control group, have comparable levels of writing proficiency and comparable levels of grasping the writing topic and applying topic knowledge. While the total posttest score of the experimental group was 3.28 points higher than that of the control group, the difference in posttest content, structure, grammar, and vocabulary scores was 0.44, 0.58, 1.58, and 0.68 points, respectively. After the independent samples t-test, their corresponding p-values were all less than 0.05, indicating that the differences in the posttest total scores and scores of each dimension between the two groups were significant. It indicates that after the teaching reform of open education, the effect of resource allocation is obvious, which prompted the students in the experimental group to improve substantially in the posttest compared with the control group.

Table 2.

Intergroup contrast

Test type	Experimental group		Control group		Independent sample T test results
Test type	M	SD	M	SD	T	P
Pretest-Total	10.53	2.036	10.56	2.176	-1.332	0.152
Posttest-Total	13.91	1.961	10.63	2.011	3.251	0.001
Pretest-Content	2.69	0.642	2.9	0.502	0.609	0.352
Posttest-Content	3.32	0.632	2.88	0.452	0.641	0.002
Pretest-Structure	2.72	0.623	2.71	0.653	0.615	0.124
Posttest-Structure	3.31	0.661	2.73	0.641	0.645	0.001
Pretest-Grammar	2.9	0.661	2.96	0.671	0.6	0.247
Posttest-Grammar	4.5	0.618	2.92	0.568	0.625	0.003
Pretest-Vocabulary	2.22	0.575	1.99	0.465	0.65	0.562
Posttest-Vocabulary	2.78	0.532	2.1	0.362	0.675	0.001

4.2.2

Paired samples t-test

The results of the paired samples t-test performed within the group are shown in Table 3. The total posttest score of the control group was 0.07 points higher than the total pre-test score, but since the sig (two-sided), i.e., p-value = 0.124 > 0.05, this difference was not statistically significant and the control group’s overall level of writing was not significantly improved. Similarly, since the P-values of the dimensions in the control group were 0.127, 0.225, 0.236, and 0.544, which were all greater than 0.05, the difference between the control group’s scores on the pre- and post-test dimensions was not statistically significant, and the control group’s students’ mastery of the topic of their writing was still skewed. And the corresponding four P-values (0.001, 0.000, 0.002, 0.001, 0.003) of the experimental group are all less than 0.05, which indicates that the students in the experimental group have made significant progress and improvement in their writing level compared with that in the pre-test.

Table 3.

Contrast results in the group

Group	Test type	Pretest		Posttest		Independent sample T test results
Group	Test type	M	SD	M	SD	T	P
Experimental group	Total	10.53	2.036	13.91	1.961	-8.332	0.001
	Content	2.69	0.642	3.32	0.632	-3.142	0.000
	Structure	2.72	0.623	3.31	0.661	-3.242	0.002
	Grammar	2.9	0.661	4.5	0.618	-3.645	0.001
	Vocabulary	2.22	0.575	2.78	0.532	-3.427	0.003
Control group	Total	10.56	2.036	10.63	2.011	-0.632	0.124
	Content	2.69	0.642	2.88	0.452	-1.252	0.127
	Structure	2.72	0.623	2.73	0.641	-1.452	0.225
	Grammar	2.9	0.661	2.92	0.568	-1.376	0.236
	Vocabulary	2.22	0.575	2.1	0.362	-1.313	0.544

4.2.3

Questionnaires

After the end of the post-test, this paper conducted an online questionnaire survey on 32 students in the experimental group using the QQ platform. The results are as follows: T1:

The search efficiency is obviously improved after using information technology to optimize the allocation of open education teaching resources.

T2:

The teaching resources allocation is rich and diverse, and the required information materials can be found adequately.

T3:

The recommended teaching resources are easy to understand and master.

T4:

The learning effect is significantly improved after adopting the optimized teaching resources for learning.

The answer options were on a 5-point Likert scale in the following order: strongly disagree, disagree, generally, agree, and strongly agree. The statistical results are shown in Figure 5. The figure shows the number of people who expressed agreement and strong agreement, and the percentage of people who agreed with the four questions was 78.1%, 81.3%, 81.3%, and 78.1%, respectively. The results of the questionnaire show that after the setting of resource sharing and personalized recommendation for open education teaching resources, the students’ learning efficiency and performance have been significantly improved, and the students are more satisfied with the teaching reform in this way.

5

Conclusion

In this paper, a resource sharing model based on blockchain technology and a resource recommendation model based on K-Means clustering algorithm are constructed as a way to optimally allocate teaching resources in open education. The constructed resource sharing model and personalized recommendation model have obvious advantages of use. After the two models are applied to optimize the allocation of teaching resources in open education, the results of independent samples t-test and paired samples t-test show that students’ performance improves significantly after the teaching reform. According to the questionnaire survey, students were satisfied with the four questions designed to support teaching reform at 78.1%, 81.3%, 81.3%, and 78.1%, respectively. It shows that the implementation effect of resource allocation optimization design in this paper is recognized by most students.

Language:: English

Publication timeframe:: 1 times per year
Journal Subjects:: Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other

Journal RSS Feed

Using Information Technology to Optimize the Allocation of Teaching Resources in Open Education and Educational Reform Practices

Kuizhong Xue

Haiyan An

Published Online: Mar 19, 2025

Received: Oct 31, 2024

Accepted: Feb 11, 2025

DOI: https://doi.org/10.2478/amns-2025-0418

KeywordsBlockchain technology, K-Means clustering, Multiple media resources, Educational reforms

© 2025 Kuizhong Xue et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Keywords
Blockchain technology, K-Means clustering, Multiple media resources, Educational reforms