Research on Innovative Curriculum System of Education and Teaching for Economic Management Majors Based on Data Visualization

Higher education is the main force of national talent cultivation, so teaching and learning in higher education is particularly important. Teaching is the most effective form of disseminating systematic knowledge and promoting students’ development, it is the basic way to carry out comprehensive development education and realize the cultivation goals, and it is the process of information interaction, thought collision and common development between teachers and students [1-3]. Curriculum teaching is the carrier to realize teachers’ professional development and students’ comprehensive development. As far as the economic management courses are concerned, there is an urgent need to build a teaching management and monitoring system, reform the teaching means and teaching methods in order to highlight the teaching characteristics of this specialty and cultivate students’ innovative thinking and practical ability [4-6]. One of the teaching objectives of economic management is to enable students to flexibly apply book knowledge to practice, which is also the practical teaching objective of this specialty [7-9]. Since in the teaching process of economic management specialties, it is necessary to focus on the development of students’ ability to participate in society and improve their personal qualities. Therefore, in the form of teaching, it is necessary to do a good job of internship and practical training, so that students can consolidate their professional knowledge in the internship work [10-11].

In the context of the big data era, data visualization technology, as an interdisciplinary course combining art and technology, design and analysis, is becoming increasingly important in its teaching status [12-13]. Data visualization technology not only involves the graphical presentation of data, but also presents complex data relationships through visual graphical charts, allowing decision makers to quickly understand the information conveyed by the data [14-16]. Data visualization technology provides a new way for the teaching reform of practical teaching courses in economics and management majors, which helps to improve the quality of teaching and the practical ability of students, promotes the close integration of education and industry needs, and takes an important step towards adapting to the educational needs of the data-driven era [17-19].

In today’s era, the cultivation goal of economic management professionals not only requires talents to master basic theories, but also to improve their quality and practical application ability, while the traditional teaching mode can no longer adapt to the requirements of the development of the times, and must be combined with modern educational technology, the innovative development of the teaching curriculum system [20]. Liu, L. et al. believe that innovation and entrepreneurship teaching reform is an imperative for the development of economics and management majors, and that simulation of the proposed innovation and entrepreneurship education model for economics and management using computers can evaluate the individual operational strategies within the teaching system in order to find a teaching innovation curriculum system that is conducive to the cultivation of innovative talents [21]. Liu, F. et al. examined the application of virtual simulation experimental teaching mode in innovation and entrepreneurship comprehensive practical training teaching of economics and management majors, and found that the change of teaching mode brought richer experimental contents as well as more diversified experimental evaluations, which helped the digital teaching reform of economics and management majors [22]. Chen, Y. emphasized that economic and management majors should focus on students’ digital thinking and digital skills training, building modular courses, put forward a “three-dimensional three-stage” modular course construction system for economic and management majors, and combined with digital tools as the implementation path to establish a teaching innovation system [23]. Yang, L. I. N. et al. optimized the teaching content and integrated simulation practical training, vocational ability practical training, scientific research and innovation practical training and entrepreneurship practical training links on the basis of analyzing the problems of practical teaching in economics and management majors to provide inspiration for the construction of practical teaching system [24]. Stanca, L. et al. showed that visualization technology plays an important role in finance teaching and learning environments by virtue of its ability to represent information in a clear and effective manner, and that modern visualization technology tools can assess student participation in classroom instruction on the one hand, and on the other hand, can deepen students’ understanding of classroom instruction by enabling them to identify the skills needed to make sound accounting decisions [25]. Mikhailutsa, O. M. et al. analyzed the role of interactive IT-based pedagogical methods for enhancing students’ professional knowledge and competence by using ant colony optimization algorithms for visualizing educational data, saving and loading graphics by simulating the dynamic network movement of a large number of ants, which not only innovated the form of education, but also significantly improved the quality of teaching [26]. Halliday, S. D. et al. found that visual teaching tools are beneficial in aiding the thinking of beginners in the course, and after investigation, it was found that the developed model visualization software improved the level of expertise of students with weak mathematical skills by providing interactive exercises and motivated students to learn about economics, enhancing the teaching and learning experience [27].

In this paper, the fuzzy C-mean clustering algorithm is improved using several methods, and the Gaussian density function is used to determine the initial clustering center, which makes full use of the density information of the data and improves the clustering effect. Density-sensitive distance is used to calculate the distance between data points, which better reflects the similarity between data points. The improved fuzzy C-mean clustering algorithm was used to visualize and analyze the data for students’ and teachers’ behavioral portraits. Subsequently, a teaching quality evaluation model based on the system dynamics model is constructed, and the effectiveness of the model is empirically examined with the example of A university, which provides a new path for the establishment of the education and teaching innovation course system.

2

Behavioral profiling methods and clustering

2.1

User profiling

To sum up, user profile is an effective tool for in-depth understanding and analysis of users, which can help enterprises better understand users’ needs and provide more personalized products and services, so as to achieve the purpose of better interaction and communication with users. With the continuous development of science and technology, the application of user profiles will become more extensive and in-depth.

User portrait modeling is the process of constructing a user profile by analyzing user data and using various data mining and machine learning techniques. The following is a general portrait modeling process: 1)

Data collection: first of all, it is necessary to collect relevant data about the user, including the user’s basic information, online behavior, purchasing habits and so on. The data can be gathered in various ways, including user registration information, questionnaires, website access logs, social media data, and more.

2)

Data cleaning and pre-processing: after data collection, it is necessary to implement cleaning and pre-processing of the original data to eliminate abnormal noise data and standardize the data format to ensure data quality and consistency.

3)

Feature selection and extraction: after data preprocessing, it is necessary to select and extract features applicable to user profile modeling.

4)

Model selection and establishment: after feature selection and extraction, applicable modeling algorithms and models, such as natural language processing, cluster analysis, deep learning, etc., can be selected according to specific needs.

5)

Model training and validation: after selecting and building the model, it is necessary to train the model using labeled training data and evaluate and tune the model using validation data. This process includes optimizing model parameters and selecting validation metrics to achieve the best model results.

6)

Model application and update: After model training and validation, the model can be applied to actual user data to classify and recommend users. Meanwhile, since the user’s features and behaviors may change over time, the model needs to be constantly updated to maintain accuracy and effectiveness.

7)

Model evaluation and improvement: After the model is applied and updated, the model effect should be evaluated based on a variety of evaluation metrics (e.g., accuracy, recall, F1 value, etc.), and the model should be continuously optimized and improved according to the evaluation results. The approximate process of user portrait construction is shown in Figure 1:

User portrait [28] modeling is an iterative process that requires the unified coordination and integration of multiple steps such as data collection, preprocessing, feature selection and extraction, model selection and building, model training and validation, model application and updating, and model evaluation and improvement. Through continuous iteration and optimization, accurate and effective user profiling models can be constructed to provide enterprises with precise user requirements and personalized services.

2.2

Fuzzy C-mean clustering algorithm improvement

Cluster analysis is an important unsupervised learning technique in the fields of data mining and statistics, which automatically divides the objects within a dataset into clusters by calculating the similarity to maximize the similarity of the objects within the clusters and minimize the differences of the objects between the clusters. This approach can help us understand the intrinsic structure of the data and discover patterns and relationships in the data, thus revealing the natural distribution of the data in the context of unsupervised learning.

Cluster analysis can discover groups of similar objects in a dataset, provide insight into underlying patterns and structures, provide a basis for decision-making, and also provide a foundation for further data mining and analysis. However, it should be noted that the results of cluster analysis can be affected by the selection of parameters, data preprocessing and other factors, and it is necessary to comprehensively consider a variety of factors in order to produce accurate and interpretable results.

2.2.1

Gaussian Density Function to Determine Initial Cluster Centers

In traditional FCM, the initial clustering centers are usually chosen randomly, and this practice may result in clustering results that are highly influenced by the choice of initial clustering centers. In order to improve this situation, a Gaussian density function [29] can be used to determine the initial clustering centers. Specifically, the following steps can be followed: 1)

Calculate the density of each data point. A Gaussian density function can be used to calculate the rate degree occupied by each data, given a student data set S = {x₁, x₂, ⋯, x_n}, where the density of each object x_i, denoted as density(x_i), is the average of the sum of Gaussian distances of the data points around x_i, and is used as a measure of the localized density in which x_i is located, and the Gaussian function is defined as shown below. (1) $d e n s i t y (x_{i}) = \sum_{j = 1}^{n} f_{G u a s s} (x_{i}, x_{j})$ (2) $f_{G u a s s} (x_{i}, x_{j}) = e^{- \frac{d {(x_{i}, x_{j})}^{2}}{2 σ^{2}}}$

where σ is the density parameter.

2)

Select the data point with the largest density as the initial clustering center. Sort all data points according to density and select the data point with the highest density as the initial clustering center. However, there is a problem with this method in that some data points may have similar characteristics and should be grouped into the same cluster, but if they are assigned to different clusters, the clustering results will be incorrect. In order to avoid that the calculated density maximum points are all in one cluster, the appropriate region length is chosen for the division and the formula is as follows: (3) $d = \frac{D_{\max}}{K}$

where the number of clustering categories K, D_max is the maximum value of the distance between two points among all data points. The algorithm for determining the initial cluster center using Gaussian density function is as follows:

Algorithm 2: Algorithm for determining the initial cluster center using Gaussian density function

Input: data set S, number of clusters K

Output: set of initial cluster centers $C = {\begin{matrix} c_{1}, c_{2}, \dots, c_{k} \end{matrix}}$

Step 1: Calculate and determine the maximum distance D_max between data points in the data space, and the constraint distance d;

Step 2: Find the densest point c₁ among all data points

Step 3: finding the 2nd densest point c₂ with distance c₁ greater than d.

Step 4: Continuously search for cluster centers of other initial clusters c₃, ⋯, c_k as per step 3;

Step 5: Return the set of cluster center vectors $C = {c_{1}, c_{2}, \dots, c_{k}}$ .

2.2.2

Density-sensitive distances

By replacing the Euclidean distance with the density-sensitive distance, the similarity and density distribution between data points can be better reflected, thus improving the accuracy of the clustering results. At the same time, this method can also avoid the defects of Euclidean distance in traditional FCM, such as the problem of sensitivity to noise and outliers. Given two data points i and j, their density-sensitive distance $D_{i j}^{ρ}$ can be calculated by the following formula: (4) $D_{i j} = \min_{p \in P_{i j}} \sum_{k = 1}^{| p |} (e^{ρ d (p_{k}, p_{k + 1})} - 1)$ (5) $D_{i j}^{ρ} = \frac{1}{ρ^{2}} l n {(1 + D_{i j})}^{2}$

computes the shortest path between two points, P_ij is the set of all paths connecting points i and j, where path p is a sequence of points consisting of a series of neighboring sample points, the length of path p is defined as the sum of the distances between neighboring points on the path, and d(p_k, p_k+1) denotes the Euclidean distance between points p_k and p_k+1; ρ > 0 is a parameter controlling the degree of sensitivity of the density, and a smoothing of the distance can be obtained by introducing ρ and scaling adjustment that makes the distance metric between points in a dense region smaller.

The modified FCM algorithm uses the density-sensitive distance as the distance metric, and its objective function is shown in Equation (6): (6) $J = \sum_{i = 1}^{n} \sum_{j = 1}^{k} u_{i j} D_{i j}^{ρ}$

Where, $D_{i j}^{ρ}$ is the density sensitive distance between sample x_i and clustering center c_j and u_ij is the affiliation of sample point x_i in subgroup j. The formula for obtaining the new affiliation matrix by minimizing the objective function while satisfying the constraints is shown below: (7) $\sum_{j = 1}^{n} u_{i j} = 1$ (8) $u_{i j} = \frac{1}{\sum_{c = 1}^{k} {(\frac{D_{i j}^{ρ}}{D_{i c}^{ρ}})}^{\frac{2}{m - 1}}}$

By obtaining $D_{i j}^{ρ}$ through the density-sensitive distance, the objective function J can be expanded as: (9) $J = \sum_{i = 1}^{n} \sum_{j = 1}^{k} u_{i j} D_{i j}^{ρ} = u_{j 1} D_{j 1}^{ρ} + u_{j 2} D_{j 2}^{ρ} + \dots + u_{j n} D_{j n}^{ρ}$

The density-sensitive distance matrix D^ρ between any two sample points is known: (10) $A = [\begin{matrix} D_{11}^{ρ} & D_{12}^{ρ} & \dots & D_{1 n}^{ρ} \\ D_{21}^{ρ} & D_{22}^{ρ} & \dots & D_{2 n}^{ρ} \\ ⋮ & ⋮ & ⋮ \\ D_{n 1}^{ρ} & D_{n 2}^{ρ} & \dots & D_{n n}^{ρ} \end{matrix}] \times [\begin{matrix} u_{11} & u_{21} & \dots & u_{k 1} \\ u_{12} & u_{22} & \dots & u_{k 2} \\ ⋮ & ⋮ & ⋮ \\ u_{1 n} & u_{2 n} & \dots & u_{k n} \end{matrix}] = [\begin{matrix} a_{11} & a_{12} & \dots & a_{1 k} \\ a_{21} & a_{22} & \dots & a_{2 k} \\ ⋮ & ⋮ & ⋮ \\ a_{n 1} & a_{n 2} & \dots & a_{n k} \end{matrix}]$

Then the clustering center c_j of class j is: (11) $c_{j} = {x_{i} | i = \arg \min A_{c j}}, j = 1, 2, \dots, k$

2.2.3

Determination of the optimal number of clusters

The number of clusters is an important parameter in clustering algorithms, which directly affects the quality of clustering results. There are many ways to determine the optimal number of clusters, such as the elbow rule and the contour coefficient method. However, there is always the problem of unstable optimal search and high computational complexity when used alone, so in this paper, we decided to use a combination of the two methods to determine the optimal number of clusters.

The contour coefficient (SC) [30] is a metric used to evaluate the clustering results, which takes values between -1 and 1, with larger values indicating more reasonable clustering results. The contour coefficient integrates the tightness within clusters and the separation between clusters, and can be used to compare the performance of different clustering algorithms on the same dataset and determine the optimal number of clusters. Usually, we will try different numbers of clusters and then calculate the corresponding contour coefficients, and finally choose the number of clusters with the largest coefficient as the optimal value. In addition, the contour coefficient can also be used to compare the performance of different clustering algorithms.

The method of calculating the profile coefficients is relatively simple, only need to calculate its a(i) and b(i) for each sample, and then substitute into the formula to get the corresponding coefficients. In the actual calculation, the iterative method is usually used to improve efficiency. Specifically, the following steps can be used to talk about the calculation: 1)

For each sample i, calculate its average distance a(i) from other samples in the same cluster.

2)

For each sample i, calculate its average distance from all other samples in the nearest cluster b(i).

3)

For each sample i, calculate its profile factor s(i).

4)

Calculate the average contour coefficient of all samples as an evaluation index of the clustering effect.

(12)

s (i) = \frac{b (i) - a (i)}{\max {a (i), b (i)}} s (i) = {\begin{array}{l} 1 - \frac{a (i)}{b (i)}, & a (i) < b (i) \\ 0, & a (i) = b (i) \\ \frac{b (i)}{a (i)} - 1, & a (i) > b (i) \end{array}

a(i) indicates the closeness of sample i to other samples in the same cluster, and b(i) indicates the separation between sample i and other clusters. The value of the contour coefficient ranges between [-1, 1], the larger the value, the better, when the coefficient is close to 1, it indicates that the clustering effect is good; when the coefficient is close to 0, it indicates that the clustering effect is average; when the coefficient is close to -1, it indicates that the clustering effect is poor.

The elbow rule is a method used to determine the optimal number of clusters in a clustering algorithm. When the dataset is divided into different clusters, the sum of squares within clusters (SSE) decreases as the number of clusters increases. Thus, when the number of clusters increases to a certain point, the rate of decrease in SSE decreases dramatically, creating an “elbow” point that corresponds to the optimal number of clusters.

The use of the elbow rule is to determine the optimal number of clusters in a clustering problem. Define the SSE formula as follows: (13) $S S E = \sum_{i = 1}^{k} | j - m_{i} |^{2}$

Where m_i is the clustering center of class i, j is a data point belonging to class i, not all points and all clustering centers to do the distance operation, otherwise the SSE will not increase with the number of clusters tend to 0.

Since it is known that SSE tends to 0 with the increase in the number of clusters, this intermediate process there will be a large rate of decline in the paragraph, and the number of clusters increases later, the rate of decline will tend to stabilize, then generally choose the end of the steepest paragraph as the true number of clusters.

The optimal number of clusters is determined by considering the contour coefficient and the elbow rule together.

3

User profiling and cluster analysis validation

3.1

User Profile Analysis

A general analysis of the student usage data from a resource perspective is presented in Figure 2, which shows the statistics of the numerical data of student data (the values of the x-axis are the results after normalization), and Figures a-d show the statistics of hotness, uploading time, popularity, and quality of the work, respectively. As can be seen from the graph, the number of “hotness” less than 0 is high, indicating that there are more resources with low hotness. More “upload time” than 0 means that there are more resources uploaded later, and to a lesser extent, there are also some resources uploaded for a longer period of time. The number of “popularity” equal to 0 is higher, and the number greater than 0 is higher than the number less than 0, indicating that most resources are still generally popular, but more popular resources are more popular than less popular ones. The “quality of work” is normal, which means that there are fewer high-quality and low-quality works, and most of the students’ works are of medium level. Therefore, it is concluded that the resources for economic management students have certain shortcomings and need to be improved in an innovative way.

The overall analysis of teacher data from the resource perspective is shown in Figure 3, which shows the statistics of the numerical type of data for teacher data (the values on the x-axis are the results after normalization), and Figures e-h show the statistics of hotness, uploading time, popularity, and quality of the work, respectively. The high number of “hotness” equal to 0 indicates that most of the resources are of average hotness. There are more “upload time” equal to 0, and a relatively large number of -1.5 and 1.5 neighborhoods, which means that there is more upload time in the middle period, and also a larger portion of recent uploads and long time ago uploads. The high number of “popularity” values equal to 0 indicates that most of the resources are of average popularity. “Quality of work” is highest around 1, with the remainder around 0, and a portion less than -2, indicating that among the teacher population, teaching resources are evaluated in absolute terms, either very well or very poorly, and that there is still some room for improvement in the overall teaching resources for teachers.

3.2

Cluster analysis validation

In this section, the clustering effect is verified through experiments. Comparison is made through two sets of data, the sample selection is made by randomly selecting 20 groups from the student clustering group, two persons in each group, which is recorded as Sample A. Similarly, 20 groups were randomly selected from the teacher clustering group, with two persons in each group, which is recorded as Sample B. 20 groups of students and 20 groups of teachers, two persons in each group, are randomly selected from 200 users, and are equally divided into two test samples, which are recorded as Sample C and Sample D. The statistical results of the number of questionnaires that interacting with sample B users are shown in Fig. 4.

The statistical results of the same number of questionnaires interacted by users of Sample A and Sample B are shown in Fig. 4. The statistical results of the same number of questionnaires interacted by users of Sample C and Sample D are shown in Fig. 5. The statistical results of the four groups of sample means are shown in Table 1. The mean value of sample A is 3.76, which is significantly higher than that of samples C and D. From the statistical results, it can be seen that the number of users interacting with the same questionnaires is higher on average than that of non-clustered groups in clustered groups, which indicates that in the same clustered group, the behavioral similarity of the users is stronger, and therefore it can be proved that the effectiveness of the results of the improved FCM clustering can be demonstrated.

Table 1.

Sample mean

Sample	Mean
A	3.76
B	3.38
C	2.76
D	2.32

4

Ideas for reforming the teaching of economic management courses

First of all, the teaching concept of economic management courses has been innovated. As we all know, talent is the core force of innovation and entrepreneurship, if colleges and universities follow the traditional backward education model, it will lead to students can not correctly grasp the concept of innovation and entrepreneurship, do not have the spirit of innovation. For example, the concept of innovation and entrepreneurship is limited to the opening of online stores, micro-business and open traditional stores and other forms, in the specific entrepreneurial process, but also due to the lack of legal awareness and do not dare to safeguard the legitimate rights and interests of individuals, and even their own will violate the law and integrity of the business. Therefore, in the teaching of economic management courses in colleges and universities, it is necessary to introduce the concept of innovation and entrepreneurship teaching into the classroom, clearly point out the qualities that entrepreneurs need to have, pay attention to guiding college students to carry out career planning, and strengthen the integrity of college students and the rule of law ideological education.

Secondly, the teaching content of economic management courses should be refined. In the context of innovation and entrepreneurship, we should pay attention to combining the economic management professional courses with innovation and entrepreneurship, and pay attention to the refinement of the teaching content. According to the personalized needs of different students, provide different teaching content, for some teachers who clearly expressed the idea of entrepreneurship should be targeted to guide the relevant personnel appointment, management system, enterprise establishment and other legal knowledge to form a system to guide the teaching.

Again, reform the teaching method. In the context of innovation and entrepreneurship, the traditional teaching methods of economic management courses in colleges and universities appear to be very backward, the reason for this is that every innovative entrepreneur needs to master the basic economic and legal knowledge, and combined with the actual problems that may be encountered in the process of innovation and entrepreneurship, the theoretical knowledge of economic management is applied in practice.

This requires accelerating the reform of teaching methods. For example, inviting successful entrepreneurs to give lectures in schools to teach successful entrepreneurial experience, or simulating entrepreneurial projects and organizing students to participate in the preparation of contracts and so on. In practical teaching, students can realize the importance of mastering economic and legal knowledge for innovation and entrepreneurship, and how to apply economic and legal knowledge in practice.

Finally, improve the ability level of the teaching team. Teachers, as people who preach and teach, play a vital role in the cultivation of innovative and entrepreneurial talents, and colleges and universities should be good at introducing high-quality teachers specialized in economic management and injecting fresh blood into the teaching team of colleges and universities. In addition, they can regularly organize economic management teachers to practice and exercise in innovation and entrepreneurship bases, improve teachers’ own practical application ability, hire successful innovation and entrepreneurs to serve as guest teachers in colleges and universities, and give full play to the benchmarking role of these successful people. At the same time, teachers who specialize in economic management are encouraged to work in law firms or innovation and entrepreneurial practice groups to acquire practical experience.

5

Evaluation of teaching quality in higher education based on system dynamics

5.1

Construction of Teaching Quality Evaluation System in Colleges and Universities

The characteristics of data visualization-based education are openness, large-scale, focus on people, subversion of operational modes, and focus on ecology. As cloud computing, mobile Internet and so on are more and more deeply rooted in the education system, the data visualization-based education model will inevitably cause profound changes in the education system and technology, reshape the new concept of education, and build a new business model of education. In recent years, many colleges and universities, education and training institutions and other institutions have actively promoted and applied data visualization-based education model, and have achieved many applications and research results, presenting MOOC, blended teaching and other diversified teaching methods. However, as an emerging thing, the education scene based on “Internet +” has significant differences with the traditional classroom teaching mode, and the teaching quality evaluation system of this mode is still imperfect at this stage, so how to comprehensively evaluate the teaching quality by cultivating the Internet education thinking, so as to improve the teaching effect and enhance the ability to educate people is an issue that is becoming more and more prominent. This problem is becoming increasingly noticeable and requires immediate attention.

Based on the data visualization education model college teaching activities as an open, complex system, involving more subjects, links, etc., the quality of teaching is obviously affected by many subjective and objective, internal and external factors. Therefore, the analysis of its influencing factors should be carried out with the main line of teaching activities as the center and teaching as the goal. Combined with the analysis of literature research, the questionnaire program is formulated. Starting from the concept of teaching systems, teaching activities involve the four most important elements of teachers, students, management, and services, which are considered as the first-level influencing factors. Based on this, the second-level influencing factors are subdivided; at the same time, the second-level factors and their subdivided elements are provided with possible expansion options. In this round of questionnaire survey, questionnaires were sent to some teachers, educational administrators and some students of several professional classes in S. Based on the questionnaire recovery, the survey results were analyzed and sorted out to reorganize the elements at all levels. Accordingly, the main influencing factors at all levels were identified. The nature and extent of the impact of each element and its related refined decomposition factors on teaching quality will be reflected in the project system model. According to the analysis of the influencing factors of teaching quality in this model of higher education, taking the four core elements as the nodes, the elements are refined and decomposed to establish the teaching quality evaluation index system of this model, which is shown in Table 2.

Table 2.

The evaluation index of teaching quality in colleges and universities

Influencing elements	Elaboration
Teacher resources	Personnel training funds,Innovate teaching fund
Students	Innovate students cultivate funds
Management	Proportion of graduate degree administrators
Service	Infrastructure capital,Innovate resources construction fund,Normal year money

5.2

SD-based teaching quality evaluation model for higher education institutions

SD believes that the source of the driving force for the development of the system comes from the elements within the system, and that any particular dynamic behavior of the system is generated by the interactions of the elements within the system boundaries, which delineate the smallest number of units that need to be included in order to form a particular behavior. Therefore, from a system perspective, the core and key influencing factors within the system boundaries are identified based on the influencing elements and their relationships from a system perspective, centered on the system goals and problem-oriented. Based on the data visualization education model of university teaching activities is “teaching” and “learning” bilateral activities, supplemented by teaching management “management” and service platform of the “Link”. Therefore, when constructing the teaching quality evaluation model based on the data visualization education model, key elements such as teaching subjects and conditions related to teaching activities are incorporated into the system, and their logical relationships and evolutionary mechanisms are analyzed and sorted out and revealed by research.

According to the above analysis, based on the principle of SD, Vensim PLE software is used to draw the stock flow diagram of the teaching quality evaluation system of colleges and universities based on the data visualization education model as shown in Figure 6. The stock flow diagram can not only clearly reflect the influencing factors of teaching quality and their logical relationships, but also reflect the nature of each variable, providing a basis for modeling simulation and analysis.

5.3

Simulation Analysis of Educational and Pedagogical Innovation Strategies

Simulation is carried out using Vensim software, the simulation interval is 2008~2023, and the simulation step size DT=1. Among them, 2008~2017 is the realistic fitting of historical years, and 2018~2023 is the trend prediction of planning years. Due to the large number of parameters in the system flow diagram, this study mainly takes the innovatization capital investment structure, annual growth of funds, the structure of the management team and the incentive policy variables as examples of modulation simulation to observe the dynamic change relationship of the level of educational and pedagogical innovatization under the combination of different strategies. First of all, according to the research data on the construction of educational innovativeness in university A, the model variables are assigned values, and according to the actual statistics, the proportion of funds invested in the innovativeness of the university in terms of the talent team, student cultivation, infrastructure, resource construction, and teaching and learning application is about 0.05: 0.05: 0.70: 0.10: 0.10, and the growth rate of the normal annual funding is about 10%, the proportion of the management personnel with postgraduate qualifications is is 0.3, and the influence factor of incentive policy is 1. Subsequently, a total of four strategy regulation programs are set up, A, B, C, and D, and the parameter values are shown in Table 3. Finally, the different strategies are simulated sequentially, and the dynamic changes of the number of innovative managers, the total value of innovative fixed assets, the number of innovative talent team, the total amount of innovative resource construction, the number of innovative teaching effective courses, the number of innovative student cultivation, and the simulation results of the level of educational innovativeness are shown in Fig. 7-Fig. 13.

Table 3.

Strategy ais the control parameter of A, B, C, D

Regulatory parameter	Strategy A	Strategy B	Strategy C	Strategy D
The talent team training fund input coefficient	0.1	0.1	0.1	0.1
Innovate students cultivate the investment coefficient of funds	0.1	0.1	0.1	0.1
Investment coefficient of infrastructure construction fund	0.3	0.3	0.3	0.3
Investment coefficient of innovate resource construction capital	0.15	0.15	0.15	0.15
Innovate teaching fund input coefficient	0.25	0.25	0.25	0.25
Annual growth rate of normal annual investment	0.10	0.18	0.18	0.2=18
Proportion of graduate degree administrators	0.3	0.3	0.7	0.7
Incentive policy on the influence factors of innovate teaching	1	1	1	1.3

The simulation results are analyzed as follows: 1)

From the simulation results, it can be seen that educational innovatization is a dynamic development process that continues to rise. In the previous period, the school invested a large amount of special funds to carry out the construction of the basic environment, including the construction of campus computer network system, digital library literature and information guarantee system, modern education technology digital comprehensive teaching support system, etc., the level of educational innovativeness rises slowly; after the financial investment, policy support and incentives, the overall level of educational innovativeness rises significantly.

2)

From the historical yearly data, it can be seen that the school did have the phenomenon of emphasizing hard over soft and unbalanced allocation of funds in the past. The simulation results of Strategy A, which reduces the financial investment in infrastructure and increases the financial investment in teaching and resources, indicate that the level of educational innovatization increases or decreases after adjusting the structure of financial investment in innovatization.

3)

Continuous capital investment is the most important prerequisite guarantee in the work of innovatization in colleges and universities. Strategy B increases the total investment of capital for innovatization and raises the annual growth rate of capital from 10% to 18%, and its simulation results show that it has a positive effect on innovatization of fixed assets, talent team construction, innovatization of teaching, students’ cultivation, and resource construction.

4)

Strategy C enhances the ratio of graduate students of innovative management personnel from the actual 0.3 to 0.7, and its simulation results show that innovative management, talent team construction, resource construction, innovative teaching, student training, and educational innovativeness are all enhanced with it.

5)

Strategy D increases the incentive policy impact factor from 1 to 1.3, and its simulation results show that it can promote innovative management, innovative teaching, student cultivation, and talent team construction, thus enhancing the level of educational innovation.

6)

The simulation results show that the elements of educational innovativeness are interrelated, and as the feedback loop listed earlier in this study, changing the value of one parameter of the system will affect multiple variables. The level of educational innovation is not affected by a single factor alone, but by a combination of multiple factors. Therefore, in the process of system development, the relationship between the whole and the local should be properly handled to achieve rational resource allocation and coordinated development.

6

Conclusion

Based on the strategic needs of modernizing education and integrating data visualization into education, this paper proposes an innovative strategy for the curriculum system of the education and teaching model of economic management.

There are more resources with low enthusiasm in the student profile, and most of the resources are generally popular among students. Students’ “quality of work” is normal, indicating that there are more medium-quality works and a lack of high-quality works. From the students’ portraits, it is concluded that there are certain deficiencies in the resources for students majoring in economics and management, and there is room for innovation and improvement in teaching.

Teachers’ portraits show that the teachers’ group’s evaluation of teaching resources is more absolute, which laterally reflects some teachers’ dissatisfaction with teaching resources, and the establishment of innovative course system can start from teachers’ teaching resources.

The simulation results of the educational teaching innovation strategy show that the elements of educational innovativeness are closely related, as the feedback loop listed in the previous section of this study, such as increasing the annual growth rate of funds from 10% to 18%, increasing the proportion of graduate students of innovativeness management personnel from 0.3 to 0.7, or increasing the impact factor of incentive policy from 1 to 1.3, which are the behaviors to change a certain parameter, will affect more than one variable. It can be concluded that the level of educational innovation is simultaneously affected by the combination of multiple factors. Therefore, in the process of establishing the innovative curriculum system of economic management professional education and teaching, the relationship between various innovative elements should be appropriately dealt with to realize the rational allocation and coordinated development of educational resources.

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

Research on Innovative Curriculum System of Education and Teaching for Economic Management Majors Based on Data Visualization

Hongkai Cui

Published Online: Mar 21, 2025

Received: Nov 04, 2024

Accepted: Feb 10, 2025

DOI: https://doi.org/10.2478/amns-2025-0594

KeywordsFuzzy clustering, User profile, Profile coefficient, Data visualization, Economic management professional education

© 2025 Hongkai Cui, published by Sciendo

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

Keywords
Fuzzy clustering, User profile, Profile coefficient, Data visualization, Economic management professional education