Design of Artificial Intelligence-Driven Multi-Level Management Strategies for Internationalization Management in Higher Education and Its Feasibility Study

In recent years, with the advancement of globalization and the rapid development of the economy of each country, the internationalization of higher education has become a hot topic of concern in the academic and educational circles. The internationalization of academic research, teaching and training, and the flow of talents has accelerated the reform and development of the higher education system in various countries [1–4]. The internationalization management of higher education not only embodies the communication and mutual influence among the fields of economy, science and technology, culture and so on, but more importantly, it provides more opportunities and challenges for the reform and development of higher education system of each country, and the application of artificial intelligence in the internationalization of colleges and universities will increase the speed of the process of internationalization management of colleges and universities [5–8].

With the continuous progress of science and technology and the rapid development of information technology, artificial intelligence technology has been widely used in various fields. In the internationalization management of higher education, artificial intelligence technology is also gradually showing its great potential and advantages [9–12]. Artificial intelligence technology plays an important role in improving the quality of education, enhancing students’ learning experience and teachers’ teaching effect. Through intelligent school management, personalized learning support and intelligent teaching assistance, it can better meet the needs of students and teachers [13–16]. However, at the same time, there is a need to face the challenges of data privacy and security issues as well as technology popularization and training, which can be solved through the establishment of perfect systems and mechanisms [17–18].

In this paper, we designed the strategies for the construction of informationization of education management in higher education institutions from the levels of enhancing the informationization ability of related personnel, building information technology team, improving the mechanism of education management informationization, and strengthening the construction of resource sharing system, etc. We designed the evaluation index system for education management according to the six first-level indexes, namely academic achievement, research ability, teaching ability, career development ability, comprehensive quality, and internationalization ability. Combining the advantages of the entropy weight method and the CRITIC method, the weights obtained based on the two methods are combined and assigned, and the multiplicative integration method is chosen to calculate the comprehensive weights. The cloud model using a specific algorithm to replace the affiliation function in the fuzzy comprehensive evaluation is proposed as a hierarchical comprehensive evaluation model to evaluate the teaching management strategy of higher education institutions, and the feasibility of the multilevel management strategy is verified.

2

Strategies for building information technology for multilevel management in higher education institutions of higher learning

2.1

Strengthening the information technology talent team

Higher vocational colleges and universities, with the support of the education department, in which the chief information officer of the school will be taken by the leadership of the school, will make use of the on- and off-campus information network and resource base to launch the construction of the information technology application training base, and with the help of the driving effect of the demonstration project, will carry out the training of the ability of multimedia technology, informatization technology and Internet application, learn and master the new working technology and methods, and train the education administrators and teachers who are closely related to informatization. The training of management personnel and teachers will ensure that they can play an important role in the smooth advancement of the process of digitizing education management.

The people who can contribute to information management are organized into teams to ensure that the information construction can be carried out smoothly; at the same time, certain declaration training needs to be carried out in the school, through which the training helps to determine the theme of the seminar accordingly, choosing the teachers to choose to learn, research areas, school-based training, etc., and the subject information technology teaching is considered to be the core of the entire construction of the information technology in teaching, and through which the relevant training for teachers’ information technology references can be promoted. Promote the development of relevant training activities for teachers’ information technology references.

2.2

Accelerate the construction of a resource sharing system

In order to solve the problems of uneven development of informatization construction in central and western colleges and universities, insufficient construction of high-quality digital educational resources and insufficient integration into teaching applications, the state has carried out the corresponding construction and supply of professional teaching resource base projects for vocational education, and has established a professional teaching resource base for 1500 majors including different colleges and universities in central and western China to a certain extent by means of the way of “common construction and sharing” of the corresponding teaching resources among different regions. Through this way of “sharing” teaching resources between different regions, a professional teaching resource base for 1500 majors has been established between different institutions in central and western China, which to a certain extent has solved the problem of insufficient investment in informatization construction of some institutions in central and western China.

Since the concept of education informatization has been put forward, all higher vocational colleges and universities have been actively conducting research and construction of information resources. By analyzing the characteristics and advantages of higher vocational colleges and universities and the types of schools they run, and by analyzing the construction of information groups by strong scientific research and teacher teams, we should try our best to make the resource library cover all majors, highlight the role of the resource library as a supplement to extracurricular self-study and classroom teaching, and pay attention to the improvement of the professional knowledge, theory and vocational skills of the students etc. to build a resource base suitable for their own school. Higher vocational colleges and universities are lagging behind undergraduate schools in educational informatization resource construction, and the scientific research ability of teachers is relatively weak compared to undergraduates, and they lack sufficient experience in information resource construction, therefore, they should refer to and learn from the successful experience of other higher vocational colleges and universities in the construction of information resources and software development, or build the information resource base together with other sister colleges and universities or organizations.

2.3

Reform the management mechanism of informatization education in higher vocational colleges and universities

The fundamental goal of education management is to improve the quality of education, but under the condition of perfect assessment system, improving the quality of education plays a very important role, and it can be seen that the two have a complementary relationship. In this regard, a perfect quality assessment mechanism must be established to ensure the informationized education management of higher vocational colleges and universities, and it is taken as the focus of the education management reform of higher vocational colleges and universities to grasp, firstly, in order to expand the scale of schooling, and secondly, to improve the quality of schooling. In addition, the lack of an overall information technology assessment system has become a very prominent problem. At present and for a long time in the future, information technology is a major development trend, and in order to be more in line with the information technology-centered development trend, higher vocational colleges and universities need to explore a road that is in line with their own development, to be in line with the information technology, to carry out the effective innovation of the education and management mode through the way of integrating the resources, to improve the informationization assessment system of education management, and to regard informationization as one of the core strategies of higher vocational colleges and universities. The informationization work should be regarded as one of the core strategies of higher vocational colleges and universities.

3

Educational management strategy evaluation system and cloud model construction

3.1

Establishment of an indicator system

Through the research, the six themes of academic achievement, research ability, teaching ability, career development ability, comprehensive quality and internationalization ability were developed to provide a framework for a diversified evaluation index system for education management. Table 1 shows the framework of the evaluation index system, in which each index is designed to comprehensively assess the overall ability of the students, from the traditional academic performance, such as grades and thesis publication, and then extended to the teaching situation, career planning, teamwork, and professional ethics. Student education should not only focus on students’ academic achievements, but also cover the multidimensional competencies of research ability, teaching skills, personal development and interaction with the society, so as to reflect and promote the comprehensive and balanced growth of students, to meet the diversified needs of student education in different fields, to help educational administrators better grasp the quality of education, and to promote the improvement of education and the professional development of individual students.

Table 1.

Evaluation index system

Primary indicator	Index project	Explanation
Academic achievement	Published quantity	The number of papers published in core journals and general journals
	Thesis quality	The presentation of the paper in the discussion of the peer-reviewed and the frequency of the frequency of the introduction
	Academic award	The number of academic prizes and their grades
	Professional test results	The achievements of the specific disciplines
Research ability	Research project participation	The number of research projects and personal roles of different levels of research involved
	Data analysis capability	The technique and efficiency of the analysis of the log
	Research innovation	The innovation and unique contribution of the research
	Research applicability	The practical application potential and effect of the research results
Teaching ability	Teaching evaluation	The evaluation of the teaching effect of students and peers
	Content update	The content of the teaching content is related to the latest research trend
	Classroom management	Effective management of class order and ability to arouse learning interest
	Guide the student study	To guide students’ ability and results in the research projects
Occupational development ability	Internship and work experience	Internship and work experience during the study period
	Occupational planning assessment	The clarity and feasibility of individual career planning
	Vocational training	Master the skills of the corresponding profession
	Employment and satisfaction	Employment and satisfaction after graduation
Comprehensive quality	Leadership and teamwork	Leadership and teamwork ability in organizational activities and projects
	Academic ethics and occupational ethics	Observe the performance of academic ethics and occupational ethics
	Cross-cultural communication	Communication and collaboration in international environment with different backgrounds
	Social responsibility and civic consciousness	Show attention to social issues and participate in improving social action
International ability	Foreign language level	Fluency and application of foreign language
	International exchange experience	Participate in the experience of international conference and exchange programs
	International cooperation project	Participate in the number and role of international scientific research cooperation
	International vision	Understanding and grasp of international academic and educational trends

3.2

Indicator empowerment

3.2.1

Entropy weight method

Entropy weight method is an objective method based on information entropy [19], which is not affected by subjective factors. And commonly used multi-criteria decision-making method, taking into account the correlation and mutual influence between different indicators, by calculating the entropy value and weight of the indicators, it can comprehensively and accurately reflect the interrelationship between the indicators, so as to better analyze the decision-making.

There are positive and negative indicators for each indicator, positive indicator is the larger the better type of the indicator, with a great value attribute; negative indicator is the larger the worse type of the indicator, with a very small value attribute. Construct the original matrix C = (c_ij)_m×n, c_ij as the initial value of the j indicator in the ith year. Normalize the original matrix by the following formula to get matrix A = (x_ij)_m×n.

Positive Indicator: 1 $x_{i j} = \frac{c_{i j} - c_{\min}}{c_{\max} - c_{\min}}$

Reverse Indicator: 2 $x_{i j} = \frac{c_{\max} - c_{i j}}{c_{\max} - c_{\min}}$

Calculate the weight of the j st indicator for year i: 3 $P_{i j} = \frac{x_{i j}}{\sum_{i = 1}^{n} x_{i j}}$

Information for calculating indicator j e_j: 4 $e_{j} = - \frac{1}{\ln m} \sum_{i = 1}^{\infty} (x_{i j} \ln x_{i j})$

Calculate weight w_j for indicator j: 5 $w_{j} = \frac{(1 - e_{j})}{\sum_{j = 1}^{n} (1 - e_{j})}$

3.2.2

CRITIC method

CRITIC method is an objective empowerment method [20], which is used in the comprehensive evaluation problem of multiple indicators and multiple objects. It eliminates the correlation between indicators and reduces the overlap of information by analysing comparative strength indicators and conflicting indicators, so as to obtain credible evaluation results. The standardization of raw data is the same as the standardization of entropy weight method, and the matrix is the same as A= (x_ij)_m×n.

After the standardization, the variability and conflict calculation and analysis are carried out for each indicator.

Calculate the indicator variability S_j, the indicator variability is presented in the form of standard deviation, the calculation formula: 6 $S_{j} = \sqrt{\frac{1}{m - 1} \sum_{i = 1}^{m} {(x_{i j} - {\bar{x}}_{j})}^{2}}$ where ${\bar{x}}_{j} = \frac{1}{m} \sum_{i = 1}^{m} x_{i j}$ . Eq. x_j is the average of the j rd indicator.

Calculate the indicator conflictivity R_j: 7 $R_{j} = \sum_{j = 1}^{n} (1 - r_{i j})$ where r_ij is the correlation coefficient of indicator j in year i.

Calculate the information quantity G_j: 8 $G_{j} = S_{j} R_{j}$

Calculate objective weight w_j: 9 ${w^{″}}_{j} = \frac{G_{j}}{\sum_{i = 1}^{n} G_{j}}$

3.2.3

Combined weights

Aiming at the respective advantages and disadvantages of the objective assignment method, entropy weight method and CRITIC method, in order to reflect the decision makers on the degree of importance of the index attributes, increase the accuracy of the assignment, the weights obtained from the two kinds of assignment method for the combination of the assignment to reduce the arbitrariness of a single assignment method, this paper selects the multiplication and integration method to calculate the composite weight, so that the decision-making results of the real, scientific and credible, the calculation formula: 10 $w_{j} = \frac{{w^{'}}_{j} {w^{″}}_{j}}{\sum_{j = 1}^{n} {w^{'}}_{j} {w^{″}}_{j}}$ where w_j is the combined weight of the j nd indicator; ${w^{'}}_{j}$ is the weight of the j th assignment method, and w″ is the weight of the j th CRITIC.

3.3

Comprehensive Grade Evaluation-Cloud Model

Cloud model is a new theory that has been refined on the basis of the concept of affiliation function, using a specific algorithm to replace the affiliation function in the fuzzy comprehensive evaluation [21], and an evaluation model that transforms qualitative concepts and quantitative expressions to make the results more scientific and reliable. It is now widely used in system comprehensive evaluation and so on. The cloud model realizes the transformation of qualitative concepts and quantitative data by means of the cloud’s digital features E_x, E_n, H_e and combining with the cloud generator. The specific algorithm steps are as follows: 1)

Build a standard cloud. According to the above definition of maturity level combined with the expert opinion obtained from the consultation will be intelligent construction application maturity level is divided into the following five levels: for the S1 level (very poor), S2 level (poor), S3 level (general), S4 level (good), S5 level (very good), at the same time, in order to reflect the differences in the various grades, using the golden section method of thesis field [0,1] to set up the score interval, the score interval as shown in Table 2.

The cloud parameters of each grade interval and the number of cloud drops N are converted by the forward cloud generator to obtain the cloud map of the maturity evaluation criteria for the application of informationization construction strategies in education management, and N = 1000 is set according to the law of large numbers in order to improve the precision and avoid the fuzzy error caused by the large randomness.

2)

Calculation of indicator cloud parameters. Three numerical features of the cloud model are calculated based on the scoring results of n expert combined with the cloud generator. The calculation formula is: 11 ${\begin{array}{l} E_{x} = \frac{1}{n} \sum_{i = 1}^{n} X_{i} \\ E_{n} = \sqrt{\frac{π}{2}} \times \frac{1}{n} \sum_{i = 1}^{n} | X_{i} - E_{x} | \\ H_{e} = \sqrt{S^{2} - E_{n}^{2}} \end{array}$

In Eq. (11): S² is the sample variance, $S^{2} = \frac{1}{n - 1} \sum_{i = 1}^{n} {(X_{i} - W)}^{2}$

3)

Calculate the first- and second-level potential indicator clouds. According to the obtained weights of the corresponding indicator combinations W_i, combined with the cloud model floating cloud algorithm, calculate the parameters of the first- and second-level potential indicator cloud model, the calculation formula is: 12 ${\begin{array}{l} E_{x} = \sum_{i = 1}^{n} W_{i}^{*} E_{x i} / \sum_{i = 1}^{n} W_{i}^{*} \\ E_{n} = \sum_{i = 1}^{n} W_{i}^{*}^{2} E_{n i} / \sum_{i = 1}^{n} W_{i}^{* 2} \\ H_{e} = \sum_{i = 1}^{n} W_{i}^{* 2} H_{e i} / \sum_{i = 1}^{n} W_{i}^{*}^{2} \end{array}$

4)

Calculate the synthesized cloud parameters. This step is to synthesize the cloud model to get the numerical features of the synthesized cloud model and calculate the similarity between the synthesized cloud and each standard class cloud δ. The formula is. 13 ${\begin{array}{l} E_{x} = \sum_{i = 1}^{n} E_{x i} W_{i}^{*} \\ E_{n} = \sqrt{\sum_{i = 1}^{n} E_{n i}^{2} W_{i}^{*}} \\ H_{e} = \sum_{i = 1}^{n} H_{e i} W_{i}^{*} \\ δ_{B} = \frac{1}{p} \sum_{i = 1}^{p} e^{- {(b_{i} - E_{x_{B}})}_{n B}^{2 / 2 E_{n}^{2}}} \end{array}$

In equation (13): n is the number of indicators, $W_{i}^{*}$ is the comprehensive weight corresponding to the indicators, b_i is the normal random number with E_x as the expectation and $H_{e}^{2}$ as the variance, p is the number of normal random numbers, and B indicates that the value is the comprehensive cloud parameter.

5)

Plotting the integrated evaluation cloud. Through the forward cloud generator, use MATLAB software to generate a comprehensive evaluation cloud diagram of the maturity of intelligent construction capability in the space of the theory domain, combine with the cloud similarity comparison analysis, and get the evaluation results of the application of education management informatization strategy in accordance with the principle of maximum affiliation.

Table 2.

Standard grade

Standard grade	Interval
Difference (S1)	[0,28]
Worse (S2)	[23,50]
Medium(S3)	[48,78]
Better(S4)	[76,88]
Good(S5)	[85,100]

4

Determination of portfolio weights based on game theory

A higher education institution that implements the educational management information technology construction strategy proposed in this paper will be randomly selected to explore its feasibility.

Comprehensive consideration of analyzing the subjective and objective weights of the school’s educational management informatization construction strategy can obtain relatively reliable indicator weights, in which how to determine the respective weight coefficients of subjective and objective is a key issue. Drawing on the equilibrium theory of game theory, the Nash equilibrium point is solved to minimize the deviation between the comprehensive weights and the subjective and objective weights, and to achieve the optimal combination of weights. Using MATLAB software to weight the obtained AHP index weights and the improved CRITIC method index weights according to the formula for the calculation of weight coefficients and comprehensive weights of the indicators, the coefficients of the first-level indexes, the subjective AHP weights of the second-level indexes, and the objective and improved CRITIC method index weights can be obtained, and finally the combined weight value of the index system can be calculated, and the combined weight value is shown in Table 3.Academic Achievements, The index weights of research ability, teaching ability, career development ability, comprehensive quality and internationalization ability are 0.111, 0.135, 0.256, 0.218, 0.152, 0.128, respectively, of which teaching ability has the highest weight proportion.

Table 3.

The education management evaluation index comprehensive weight

Primary indicator	Weighting	Secondary indicator	Weighting
Academic achievement	0.111	Published quantity	0.028
		Thesis quality	0.025
		Academic award	0.025
		Professional test results	0.033
Research ability	0.135	Research project participation	0.024
		Data analysis capability	0.028
		Research innovation	0.036
		Research applicability	0.047
Teaching ability	0.256	Teaching evaluation	0.058
		Content update	0.062
		Classroom management	0.057
		Guide the student study	0.079
Occupational development ability	0.218	Internship and work experience	0.044
		Occupational planning assessment	0.057
		Vocational training	0.062
		Employment and satisfaction	0.055
Comprehensive quality	0.152	Leadership and teamwork	0.035
		Academic ethics and occupational ethics	0.042
		Cross-cultural communication	0.021
		Social responsibility and civic consciousness	0.054
International ability	0.128	Foreign language level	0.022
		International exchange experience	0.036
		International cooperation project	0.024
		International vision	0.046

Comparing and analyzing the AHP and improved CRITIC method weights of the secondary indicators and the final combined weights, the results are shown in Table 4, which shows that the subjective weights of 11 secondary indicators are higher than the combined weights, the subjective and objective weights of 1 secondary indicator are equal, and the subjective weights of the remaining 12 secondary indicators are lower than the combined weights. It can be seen that the combination of game theory-based weighting plays a role in regulating the subjective and objective weights, and the combination of weighting integrates the advantages of subjective and objective weighting, which is more accurate and credible in comparison.

Table 4.

The subjective and objective weight of the index

Secondary indicator	AHP weight	CRITIC weight	Composite weight
Published quantity	0.035	0.018	0.028
Thesis quality	0.042	0.009	0.025
Academic award	0.036	0.016	0.025
Professional test results	0.024	0.039	0.033
Research project participation	0.018	0.036	0.024
Data analysis capability	0.022	0.034	0.028
Research innovation	0.045	0.032	0.036
Research applicability	0.032	0.056	0.047
Teaching evaluation	0.058	0.058	0.058
Content update	0.078	0.051	0.062
Classroom management	0.035	0.069	0.057
Guide the student study	0.044	0.094	0.079
Internship and work experience	0.056	0.041	0.044
Occupational planning assessment	0.063	0.045	0.057
Vocational training	0.045	0.074	0.062
Employment and satisfaction	0.059	0.039	0.055
Leadership and teamwork	0.045	0.031	0.035
Academic ethics and occupational ethics	0.031	0.048	0.042
Cross-cultural communication	0.013	0.025	0.021
Social responsibility and civic consciousness	0.058	0.048	0.054
Foreign language level	0.016	0.035	0.022
International exchange experience	0.022	0.041	0.036
International cooperation project	0.018	0.032	0.024
International vision	0.059	0.041	0.046

5

Feasibility assessment of cloud-based modeling for instructional management strategies

The cloud model parameters of the second-level indicators are combined with the comprehensive weights of the second-level indicators to calculate the cloud model parameters of each first-level indicator, and based on the cloud model parameters, the cloud diagrams of the first-level indicators are generated, and the cloud diagrams of the six first-level indicators, namely, academic achievement, research ability, teaching ability, career development ability, comprehensive quality, and internationalization ability, are respectively shown in Figs. 1-Fig. 6. From the figure, it can be seen that: academic achievement A1, teaching ability A3, career development ability A4 and internationalization ability A6 are located between S4 and S5 levels, the rest of them are located between S3 and S4 levels, and the one with the highest evaluation scores is the teaching ability, whose evaluation scores are within the range of 77-94 points. In order to further verify the accuracy and reliability of the assessment results, the cloud diagram of indicators was compared and analyzed with the research results, and it was found that the cloud diagram of indicators and the research results were basically in line with each other, and that the implementation of the informationization strategy of education management had made the students’ competence in all aspects rise to the excellent level, and there was still room for improvement in the academic achievement and comprehensive quality of individual students.

Based on the comprehensive weights and the cloud model parameters of the first-level indicators, the comprehensive indicator cloud parameters are calculated, according to which the comprehensive indicator cloud of the teaching management strategy is drawn, and the comprehensive indicator cloud is shown in Figure 7. The discrete degree, cloud thickness and span range of the comprehensive assessment cloud are larger than the standard cloud, indicating that there are cognitive differences in the assessment results of different experts on the teaching management informatization strategy, which is in line with the logic of natural thinking and reflects the ambiguity and randomness of the assessment process. The impact of the school’s educational management informatization on students’ overall competence is between S4 and S5 levels closer to S5, with evaluation scores in the range of 75-94 points, indicating that its level of impact on students’ overall competence is excellent. The research found that after the implementation of the informatization strategy of education management, the students of the university have improved their performance in all aspects of competence, the overall quality level of students is high, and the information management work is in place, and it is again verified that the evaluation results basically match the research results, which further illustrates the feasibility of the design of the AI-driven multilevel management strategy in the internationalization management of higher education.

6

Conclusion

In order to test the feasibility of the multilevel management strategy constructed in this paper for higher education management, the feasibility of strategy implementation was evaluated in a university as the research object, and the evaluation results were in line with the research results, which verified the reasonableness of the construction of the evaluation model in this paper.

The four evaluation indicators of academic achievement, teaching ability, career development ability and internationalization ability are all located between very good and good grades, and the rest of the evaluation indicators are located between good and medium grades, which proves that the implementation of the strategy of informationization of education management has a certain effect on the enhancement of the students’ abilities in all aspects. And the indicator cloud map is consistent with the research results, which again validates the feasibility of implementing an educational management IT strategy.

The comprehensive indicator assessment level of the school’s educational management informatization strategy is between very good and good levels and closer to very good, indicating that the implementation of the strategy has an overall excellent impact on students’ competence. The cloud diagram of the comprehensive indicators also matches well with the research results, which further illustrates the accuracy of the assessment model, and also shows that the implementation of multi-level management strategies in higher education institutions can fundamentally improve the quality and effectiveness of student management education.

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

Design of Artificial Intelligence-Driven Multi-Level Management Strategies for Internationalization Management in Higher Education and Its Feasibility Study

Yue Wang

Jing Gao

Jia Su

Published Online: Mar 19, 2025

Received: Oct 20, 2024

Accepted: Feb 09, 2025

DOI: https://doi.org/10.2478/amns-2025-0440

KeywordsCombination Empowerment, Entropy Weight Method, CRITIC Method, Education Management Informatization, Cloud Modeling

© 2025 Yue Wang et al., published by Sciendo

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

Keywords
Combination Empowerment, Entropy Weight Method, CRITIC Method, Education Management Informatization, Cloud Modeling