Research on the Intelligent Optimization Method of Digital Communication Path of Ice and Snow Sports and Cultural Elements

Northeast China has a large number of ethnic minorities, which has given birth to more than one hundred ethnic traditional sports, among which traditional ice and snow sports are the ice and snow folk sports reflecting the unique natural ecological resources of Northeast China, as well as the ethnic minorities’ regionality, ethnicity and culture [1–2]. However, due to the geographical environment, production and life style, and religious beliefs of ethnic minorities in Northeast China, the conditions and space for the survival of ice and snow folk sports programs have gradually disappeared, restricting the development of ice and snow folk sports [3–4]. In the post-Winter Olympics period, there is a better opportunity for the cultural inheritance and development of ethnic minority ice and snow folk sports in Northeast China [5].

The high-speed development of integrated media technology, the rapid improvement of regional economy, the innovative development of ice and snow sports industry, and the state attaches great importance to ice and snow folk culture, make the cultural form of ice and snow folk sports of ethnic minorities in Northeast China active again [6–7]. How to show the ice folk sports culture of ethnic minorities on the modern Olympic and other cultural interaction platform “using sports as a bridge across countries and time and space”? How to show the humanistic value and spirit of Chinese traditional ice and snow sports on the basis of cultural self-awareness and cultural confidence? These are all topics worth studying [8–10].

The promotion of urban ice and snow sports in the digital era is different from the promotion and use of a single sports product, and it is a relatively new type of sports programs, products, concepts, cultures, etc., and its development influences involve a wide range of complex factors [11–13]. By analyzing relevant literature on urban ice and snow sports and interviewing relevant experts, we find out the problems affecting the construction, design, promotion and development of urban ice and snow sports, and summarize the factors affecting the development of urban ice and snow sports by combining the characteristics of the digital era [14–15].

This paper firstly studies the digital communication theory and strategy of ice and snow sports and cultural elements, and then elaborates the information dissemination mechanism of SIR, in order to solve the defects of the traditional SIR infectious disease model and its related models which can not match the real dissemination situation on the discrete time axis, and then proposes a discrete information dissemination model SPIR based on the potential dissemination node, and then analyzes the change rule of the node’s group in various situations, and constructs the corresponding set of discrete dissemination equations of the model. After that, the group rule of the node in various situations is analyzed, and the corresponding discrete propagation equations of the model are constructed. Finally, the digital dissemination paths of ice and snow sports and cultural elements are visualized and analyzed, and the group paths that trigger users’ high cultural information dissemination behaviors are explored.

2

Theory and strategy of digital communication of ice and snow sports and cultural elements

2.1

Theory of digital communication of ice and snow sports and cultural elements

With the popularization and development of Internet technology, new media and other social network information dissemination methods are becoming more and more common. The influencing factors of the dissemination of ice and snow sports and cultural elements based on social networks are mainly the popularity of social networks, the number of people using social network platforms, the knowledge level level of people using social networks, the number of people in each network node, the time and means of information dissemination, the source of information, the type of information and its importance, and so on.

The study on the communication characteristics and process of ice and snow sports and cultural elements based on social networks found that the flow of ice and snow sports and cultural elements in different social networks has the characteristics of relative stability, fast speed of communication, reversible direction of communication, repeated communication and reverse communication, etc., and the influence of the spreading of information has continuity.

Social networks as a medium for information dissemination is to social networks as a medium for the dissemination of ice and snow sports and cultural elements, the residents through the symbols and signals for the transmission of information, reception and feedback activities, is the exchange of opinions, ideas, emotions, in order to achieve mutual understanding and influence of the process.

Based on the study of the dissemination scale of ice and snow sports and cultural elements in social networks, the matrix Y of the social network nodes influencing the scale of the number of people is constructed as follows: 1 $\begin{array}{l} Y & = | \begin{matrix} x_{11} & \dots \dots & x_{1 j} & \dots \dots & x_{\ln} \\ X_{i 1} & \dots \dots & x_{i j} & \dots \dots & x_{i n} \\ X_{m 1} & \dots \dots & X_{m j} & \dots \dots & X_{n n} \end{matrix} | \\ (i = 1, 2, \dots \dots, m; j = 1, 2, \dots \dots, n) \end{array}$ where x_ij denotes the number of nodes in row i, column j (denoting layer j in row i).

Social Networks Total Influence Attendance Size G = ∑x_ij (i = 0,1,⋯⋯,n).

The cross-influence between the nodes of the network from the source down is expressed as the multiplication of two matrices Y, i.e. Y×Y.

In the study of evaluation of communication effect and communication impact of ice and snow sports and cultural elements based on social network, PDCA quality management method is used to detect and control the communication of ice and snow sports and cultural elements on social network platforms to improve the quality of communication. In order to quantify the dissemination effect of ice and snow sports and cultural elements based on social network platforms, a comprehensive evaluation system of ice and snow sports and cultural elements based on the dissemination effect of social network platforms is constructed to evaluate and test the information dissemination effect of different dissemination channels, to monitor the scale of dissemination of ice and snow sports and cultural elements and their dissemination effect, and to ensure the quality of information dissemination.

The population’s production and lifestyle, mental health and life safety, learning styles and content, and consumer attitudes are all affected by the dissemination of cultural elements of ice and snow sports. These influences will indirectly affect socio-economic development.

2.2

Digital communication strategy of ice and snow sports and cultural elements

2.2.1

Strengthening relevant theoretical research

The cultural elements of ice and snow sports have various forms of expression and rich content, which are gradually formed in the process of social development. The cultural elements of ice and snow sports have strong local characteristics due to historical and regional reasons, therefore, the research on myths, legends, astronomy, history, geography, arts and mathematics, calendars, nature, food and drink, festivals and other activities can be carried out to enrich the heritage and connotations of the cultural elements of ice and snow sports, and to deepen the public’s understanding of the cultural elements of ice and snow sports. Enhance the public’s understanding of the cultural elements of ice and snow sports, and enhance people’s sense of cultural belonging and self-confidence.

2.2.2

Establishment of a digital material bank for ice and snow sports culture

At present, the network publicity of ice and snow sports and cultural elements is mainly based on news reports, and most of these news are in the form of text and pictures to show the festivals and recreational activities carried out by ice and snow sports and cultural elements. Utilizing computer multimedia technology and network technology, digital materials such as pictures, cartoons, animations, micro-videos and other digital materials can be produced in the form of graphic, text, sound and image to display and publicize the elements of ice and snow sports culture, so as to expand the audience of the publicity of the elements of ice and snow sports culture.

2.2.3

Full utilization of web-based communication channels

Ice and snow sports culture elements related information can be published in news portals, government websites and other network platforms, users through hyperlinks to visit the relevant pages at any time and any place to read, understand, this dissemination behavior is based on the user has an intrinsic need to understand the relevant information, take the initiative to visit. The dissemination of culture can be infected and inculcated by creating a cultural atmosphere, so that the cultural elements of ice and snow sports are understood and accepted in a subtle way, and the development of social networks provides a new channel for the dissemination of the cultural elements of ice and snow sports. Social network is a form of information dissemination based on acquaintances, simulating the characteristics of interpersonal communication, with two-way communication, and timely feedback, high frequency of interaction, can realize the center of dissemination, key point dissemination, chain dissemination, dandelion dissemination and other forms of information sharing, the use of the social network platform, everyone is the recipient of information, everyone is also a disseminator of information, the process of dissemination, the information in the form of fission to affect more audiences. In the process of dissemination, the information affects more audiences in the form of fission, and the use of social networks for the dissemination of ice and snow sports and cultural elements will be more conducive to the realization of interpersonal dissemination, group dissemination and mass dissemination of information, so as to improve the effect of dissemination.

3

Social network information dissemination model construction

3.1

Complex network-based information dissemination models

Complex network theory is an effective method to identify and analyze mobile social network systems, which is a scientific analysis of empirical theories or facts related to complex systems. The research on the dynamics model of mobile social network information dissemination systems based on complex networks is currently a hot topic. Microblogs, which are based on weak relational network connections, and WeChat, which is based on strong relational network connections, are increasingly popular in the current study of information dissemination in complex networks. The advantage of weak relational network connection is that the information dissemination between users can be unidirectional, and unidirectional dissemination is faster and more efficient, with small-world and scale-free characteristics, which is compatible with the theory of six degrees of separation.

Based on this, experts and scholars divide information dissemination into three modes: chain type, single key point type, and center type. Chain type information dissemination is smaller in scope, single key point type information dissemination is dominated by speed, and center type information dissemination has more influential users, a wider dissemination scope, and higher influence. The advantage of strong relational network connections is that users are familiar with each other and trust each other, which leads to more efficient information dissemination.

3.2

Influences on information dissemination based on complex networks

Factors such as network structure, information content, and point in time may lead to changes in the network structure of information dissemination in mobile social networks, which may have a significant impact on information dissemination. In this paper, we study the influence factors of information dissemination based on complex networks at two levels: evolution time and network structure attributes. 1)

Evolutionary Time Influences

The change of information dissemination network structure is not static but dynamic. Such as the online active time of users,

Unexpected events can cause uncertain effects on the structure of information dissemination networks. According to the statistical analysis, the weekly active period of users is usually on Fridays, Saturdays, and Sundays, and the daily active period is on the eve of lunch, dinner, and sleep. According to this statistical analysis, it can be found that the peak period of users’ activity is concentrated in the rest period, and during the peak period, the information dissemination network structure evolves relatively fast. And as a breaking news that affects the change of information dissemination network structure, it generally goes through three processes from the appearance to the end: the starting period, the peak period, and the decline period. According to the research of previous data, it is known that soon after the occurrence of unexpected events, users’ attention to information will surge in a short period of time, the evolution of information dissemination network structure speeds up, and the efficiency of information dissemination increases.

2)

Network structure attribute influencing factors

From the perspective of complex network as a whole, the scale-free, small-world characteristics of mobile social network information dissemination network structure evolution in the existence of the characteristics of the optimal connection, that is, the network node user friends, followers, and the number of users and new nodes of the network node user connection probability shows a positive relationship. The connection probability P(k_i) is shown below: 2 $P (k_{i}) = \frac{k_{i}}{\sum_{j = 1}^{n} k_{i}}$ k_i denotes the degree value of the user node i. In addition to the merit connections, there are also random connections in the network structure, and merging the merit connections with the irregular connections, the probability of the new node user connecting with the original node user, P(k_i), is as follows: 3 $P (k_{i}) = \frac{(1 - b) (k_{j} + a_{i}) + b}{\sum_{j = 1}^{n} [(1 - b) (k_{j} + a_{i}) + b]} b \in [0, 1]$

Equation (3) in which a_i denotes the attraction influence factor and b denotes the controllable parameter, the size of the value depends on the weight of optimal connection and random connection, according to the formula, it can be seen that the size of the probability of the new node user connecting with the node user, P(k_i), is affected by the attraction influence factor, a_i, the controllable parameter, b, and the degree value of the user’s node, i, and other factors. Locally from the complex network, it is affected by common neighboring node users, common attributes of nodes, and gold-plating effect.

Influence of common neighboring node users. According to the actual investigation and analysis, it can be seen that whether the unfamiliar node users can establish effective contact between them depends largely on the common neighboring node users between them, the more common neighboring node users, the easier it is to establish contact. The more common the neighboring node users are, the easier it is to establish a connection. The relationship between node users and common neighboring node users in mobile social networks is known as a sub-network, and the number of sub-networks directly correlates with the efficiency of information dissemination.

Influence of nodes’ common attributes. In mobile social networks, unfamiliar node users with the same attributes (e.g., common behavioral habits, hobbies, ideals, etc.) are prone to form a network relationship. And this network relationship can promote the connection between unfamiliar node users, so that information can be spread effectively.

The influence of the gold-plating effect. The gold-plating effect refers to the utilization of the power of key node users in the mobile social network structure, so as to increase their own influence, and then increase the probability of linking with unfamiliar user nodes. The more active the key nodes are and the more attractive the information they provide, the higher the chance of the gold-plating effect, i.e., the gold-plating effect is positively related to the activity of the key nodes and the attractiveness of the information. At the same time, the efficiency of information dissemination is also proportional to the effect of gold plating.

3.3

SIR-based discrete information propagation models

3.3.1

SIR model

The SIR model contains three nodes [16], the susceptible population node S, the infected population node I and the immunizer node R. S(t) represents the number of susceptible people at the t time, I(t) represents the number of infected people at the t time, and R(t) represents the number of immunized people at the t time. a represents the probability that a susceptible individual is infected per unit of time, and b represents the probability that a susceptible individual becomes immune per unit of time. The SIR model is shown in Figure 1:

Assuming that the total number of people investigated during an infectious disease outbreak is N, N is assumed to be a constant, i.e., not counting the interference of population movements, deaths, births, etc. Obviously: 4 $S (t) + I (t) + R (t) = N$

The number of new infections per unit of time is positively correlated with the number of infected people I(t) and the number of vulnerable people S(t). Therefore, the number of new infections per unit of time is: 5 $\frac{d I (t)}{d t} = a S (t) I (t) - b I (t)$

Ditto: 6 $\frac{d S (t)}{d t} = - a S (t) I (t)$ 7 $\frac{d R (t)}{d t} = b l (t)$

Assuming I₀ = I(0), S₀ = S(0), where each initial value is usually set to I₀ > 0, S₀ > 0, and R₀ may be equal to 0 may be greater than 0, the normal case is set to R₀ = 0, which can be concluded after collation:

If S₀ > b/a, then the value of I(t) will gradually rise to S(t) = b/a, where it becomes the maximum value, and then will gradually decline until infinitely close to 0, and easy to infect the crowd number S(t) monotonically decreasing to a real number greater than 0.

If S₀ ≤ b/a, then S₀ ≤ b/a will keep decreasing until it tends to 0, while the number of susceptible people S(t) keeps decreasing to some real number greater than 0. Therefore, when S₀ > b/a, the infectious disease will lead to an increase in the number of infected patients, resulting in further deterioration of the disease and the spread of the epidemic. On the contrary, when S₀ ≤ b/a, the number of infected patients gradually decreases and the epidemic can be well controlled.

3.3.2

Discrete information propagation model based on potential propagation nodes

In order to solve the problem that the traditional SIR infectious disease model and its derived model cannot match the real propagation situation on the discrete time axis, this paper proposes a discrete information propagation model SPIR based on potential propagation nodes [17]. Since the traditional propagation model produces the phenomenon of repeated propagation when performing discrete solving, this paper first proposes a potential propagation node. After that, the group change rule of potential propagation nodes in multiple situations is analyzed, and the corresponding discrete propagation equation set model is constructed.

In order to avoid the previously mentioned problem of ΔI₊ being double-counted, this chapter proposes the potential propagation node, a node in a special state, whose specific concept is shown in Definition 1.

Definition 1

Susceptible nodes with infected nodes in their neighbors of potentially propagating nodes are potentially propagating nodes.

Definition 2

A susceptible node (S node) is a node that has not been informed of the corresponding information and is therefore in a state where it can receive information.

Definition 3

An infected node [18] (I node) is a node that knows the corresponding information and has the ability to disseminate that information and hence is in the state of disseminating the information.

Definition 4

An immune node (R node) is a node that is aware of the corresponding information but has lost interest in spreading the information and is therefore in the state of being immune to the information.

According to the above three types of nodes and based on the propagation mechanism of the traditional SIR model, this paper summarizes the following information propagation rules: 1)

If a susceptible node comes in contact with an infected node, then that susceptible node will become infected with probability β (called infection probability) per unit time.

2)

If an immune node comes in contact with an infected node, the immune node will not be infected either.

3)

An infected node will become an immune node by losing interest in the propagated information with probability γ (called immunity probability) per unit of time.

Definition 5

ΔI₊ is the set of all nodes that will be transformed from S nodes to I nodes at the next unit moment due to information propagation behavior.

As can be seen from Definition 5, ΔI₊ and ΔI₊ (i.e. dI(t)/dt) are completely different concepts. In fact, ΔI₊ is only the growth part of the components of ΔI. The reason for giving a new definition to ΔI₊ alone in this paper is that the way ΔI₊ is calculated in the SPIR model is quite different from that in the traditional SIR-derived model.

In this paper, a closed group of N individuals (meaning that the number of individuals does not change) is considered as a social network, where these individuals can be represented as nodes of the network, and the contact relationships between individuals can be represented as connected edges in the network. In addition, denoting the number of susceptible nodes, infected nodes, immune nodes, and potentially transmissible nodes in the network by S, I, R, and PS respectively, we have S + I + R = N. In addition to this, ❬d❭ and cc are used to denote the average degree value and average aggregation coefficient of the network, and β and γ are used to denote the transmission probability and immunization probability.

For an PS node, there may be one or more I nodes in its neighbors, and according to the mean field theory, it can be derived that the expectation of the number of I nodes present in the neighbors of a PS node is: 8 $I_{P S} = 1 + \frac{(〈 d 〉 - 1) (I - 1)}{N - 2}$

Replace I(t)S(t) with PS, then ΔI₊ should be calculated as: 9 $Δ I_{+} = P S * (1 - β')$

Based on the propagation probability β and the expectation I_PS of the number of I nodes linked by PS nodes, the expression of β′ can be obtained as: 10 $β' = {(1 - β)}^{I P S}$

Combining Eq. (9) and Eq. (10) gives: 11 $Δ I_{+} = P S * [1 - {(1 - β)}^{I_{P S}}]$

Based on the set of propagation equations of the classical SIR model, the following set of propagation equations underlying the SPIR model is constructed in this paper. 12 $\frac{d S}{d t} = - Δ I_{+}$ 13 $\frac{d I}{d t} = Δ I_{+} - γ I (t)$ 14 $\frac{d R}{d t} = γ I (t)$

The detailed expression for ΔI₊ is obtained as: 15 $Δ I_{+} = P S * [1 - {(1 - β)}^{1 + \frac{(〈 d 〉 - 1) (I - 1)}{N - 2}}]$

After obtaining the basic set of SPIR propagation equations, this paper discusses to analyze the state change of PS nodes (i.e., dPS/dt) during the information propagation process, which is shown in Fig. 2.

1)

S → I event (propagation process)

(1)

All newborn I nodes are transformed from PS nodes.

A node labeled PS → 1 indicates that it was a PS node at the previous moment and due to the propagation event it transformed into a I node at the current moment. Nodes not marked with an arrow indicate that they have not undergone a state change. It is more obvious that all the newly generated I nodes are transformed dichotomously from PS nodes and therefore have: 16 ${\frac{d P S}{d t} |}_{1} = - Δ I_{+}$

(2)

A newborn I node causes new PS nodes to be created in its neighborhood.

Similarly, a newborn I node causes the state of its neighbors to change. Nodes in these neighbors that were in S state are transformed into S nodes, while nodes in other states are unaffected. Since there is at least one I node in the neighborhood of these nascent I nodes (otherwise they would not have been PS nodes at the previous moment), and the density of pure S nodes without PS attributes in the whole network is (S – PS)/(N – 2), the expectation of the S nodes that will be affected by these nascent I nodes and change their state is: 17 $P S_{N_{+}} = (〈 d 〉 - 1) \frac{S - P S}{N - 2}$

However, it is important to note that since there must be at least one I node in the neighborhood of these newborn I nodes in the previous moment, there will not be any pure S node in the neighborhood of these I nodes that does not contain the PS attribute. Considering the average aggregation factor cc of the network, it can be concluded that there will be an average of cc*(〈d〉 – 1) nodes in the neighborhood of these nascent I nodes that have links with other I nodes. Since no pure S states will exist in these nodes, this component needs to be removed from Eq. (17), which scales as: 18 $P S_{N_{-}} = c c * (〈 d 〉 - 1) \frac{S - P S}{N - 2}$

There are links between the S → PS node and the two nascent I nodes, a situation that makes the above analysis of the change in PS nodes flawed. By derivation, the expectation of the number of nascent I nodes to which a pure S node without a PS state would link is obtained as: 19 $P S_{I_{k}} = 1 + \frac{(〈 d 〉 - 1) (Δ I_{+} - 1)}{N - 2}$

Combining Eqs. (17) to (19) yields: 20 $\frac{d P S}{d t} I_{2} = Δ I_{+} (1 - c c) (〈 d 〉 - 1) \frac{S - P S}{N - 2 + (〈 d 〉 - 1) (Δ I_{+} - 1)}$

2)

I → R event (immunization process)

The immunization event causes some I nodes to be transformed into R nodes, thus losing the PS attribute to become pure S nodes, and the immunization process is shown in Figure 3.

A total of γI nodes will be transformed into R nodes, thus affecting their γI〈d〉 neighboring nodes, which again can be obtained since they will only affect the PS nodes in these neighbors: 21 ${\frac{d P S}{d t} |}_{3} = - γ I 〈 d 〉 \frac{P S}{I + P S + R} {(1 - \frac{I}{N - 1})}^{〈 d 〉 - 1}$

Based on the analysis of the state change of the potential propagation node, the state change equation of the PS node can be obtained as: 22 $\begin{array}{l} \frac{d P S}{d t} & = - \frac{γ I 〈 d 〉 P S}{l + P S + R} {(1 - \frac{l}{N - 1})}^{〈 d 〉 - 1} \\ + \frac{Δ I_{+} (〈 d 〉 - 1) (1 - c c) (S - P S)}{N - 2 + (〈 d 〉 - 1) (Δ I_{+} - 1)} - Δ I_{+} \end{array}$

In this paper, N is used to replace N – 1 and N – 2 appearing in the above equations. The discrete information propagation model SPIR differential equation system based on potential propagation nodes is constructed as: 23 $\frac{d S}{d t} = - Δ I_{+}$ 24 $\frac{d I}{d t} = Δ I_{+} - γ I (t)$ 25 $\frac{d R}{d t} = γ I (t)$ 26 $\begin{array}{l} \frac{d P S}{d t} & = - \frac{γ I 〈 d 〉 P S}{1 + P S + R} {(1 - \frac{l}{N})}^{〈 d 〉 - 1} \\ + \frac{Δ I_{+} (〈 d 〉 - 1) (1 - c c) (S - P S)}{N + (〈 d 〉 - 1) (Δ I_{+} - 1)} - Δ I_{+} \end{array}$ where ΔI₊ is: 27 $Δ I_{+} = P S * [1 - {(1 - β)}^{1 + \frac{((d) - 1) (l - 1)}{N}}]$

Assuming that the information propagation process has only one infected node at the very beginning, the initial node state situation in the network is S(1) = N – 1, I(1) = 1, R(1) = 0, PS(1) = 〈d〉.

The set of SPIR information propagation equations constructed in this paper solves the problem of double counting of AA in the discrete solution of traditional SIR and its derived models. Based on the potential propagation nodes, the model analyzes the node state changes during the propagation process from the perspective of discrete time, which avoids the situation where susceptible nodes will be double-counted as infected nodes.

4

Analysis of information dissemination networks

4.1

Analysis of communication efficiency

The adjacency matrix of information exchange of network opinion leaders was constructed based on whether or not the network opinion leaders interacted with each other. And the visualization of the network was presented by Gephi software, and the calculation part of the measurement network index was carried out by Ucinet software. Based on the information interaction between network opinion leaders, the overall network diagram of network opinion leaders was constructed, and part of the network diagram is shown in Figure 4. The circular box represents 230 network opinion leaders on the topic of ice and snow sports and cultural elements, the same color individuals in the figure have similar frequency of information interaction, while the size of the shape in the figure represents the size of the centrality of the degree of the individual, and the line in the figure represents the relationship ties of network opinion leaders seeking advice or information interaction. In Figure 4, it can be clearly seen that there is no isolated node in the whole cultural communication network, indicating that there is no information silo in the 230 network opinion leaders of the ice, snow and sports culture element topic community, and all the information can be reached. After constructing the overall network diagram, it is necessary to evaluate the communication efficiency of the network through relevant indicators. The main indicators used are overall network density, shortcut distance, and network cohesion. The overall density of the network is 0.0425, which is far less than 1, and the overall density is low, which also indicates that the opinion leaders are loosely connected in the information dissemination network. Since the shortcut distance of the overall network is calculated to be 2.899, it means that each individual passes through an average of three people in this network before he or she can communicate with other individuals, resulting in a less efficient network dissemination. In addition, the cohesiveness of the network is 0.245, which indicates that the overall network is loose, and the degree of tightness and propagation efficiency between individuals needs to be improved.

4.2

Analysis of dissemination pathways

The communication path is a quantitative analysis of the degree of centrality of individuals on the network, and identifies the key nodes on the communication path of the topic from the depth of influence, which is mainly measured by the network centrality in the social network analysis method. Network centrality indicators mainly include degree centrality, middle centrality, and near centrality. Degree centrality represents the number of connections with other nodes and is divided into two categories: point-in and point-out. Part of the measurement results are shown in Table 1, in the opinion leader information exchange network the higher point-in degree is individual 14, 139, that is, the above individual is good at receiving information from other users. Higher point out degrees are individuals 53, 57, 188, who are willing to actively discuss topics with other individuals and are good at expressing their opinions. Meanwhile, intermediate centrality measures the extent to which network nodes have control over network resources. Individuals 14, 26, 53, 63, 139, 167, 188, 219, who have high intermediate centrality in the middle of the network, are the socializers of the topic community, i.e., there are more individuals in the network who need to pass through it in order to happen to be in contact with other individuals, and they also influence the dissemination of the information and knowledge related to the elements of the ice, snow, and sports culture. And according to the analysis of proximity centrality, it is found that individuals 53, 57 and 117 have higher proximity centrality points out, these individuals have more contact with other individuals in the network, are good at expressing their opinions and are less influenced by the information of intermediaries. Therefore, through the analysis of the indicators of network centrality, it is found that a larger number of network opinion leaders can access more information in the network, and are the social or information leaders of the whole network, and different types of other users with different characteristics obtain information through the interactive behaviors with them, and the information dissemination paths of the whole network increase, and no longer those opinion leaders with absolute control status play a decisive role for the information in the network. It is no longer the opinion leaders who have absolute control over the information in the network that play a decisive role.

Table 1.

Network opinion leader network center analysis

Number	Degree of degree		Intermediate center	Proximity center
Number	Point of entry	Point of point	Intermediate center	Point of entry	Point of point
4	32	24	931.468	1.142	42.44
7	24	5	-0.002	1.155	0.458
14	39	27	1329.819	1.137	38.833
20	20	0	0.001	1.151	0.453
26	15	55	2250.844	1.142	48.665
30	20	40	1008.741	1.138	46.299
35	3	45	659.182	1.138	44.784
39	29	37	1100.43	1.142	44.601
40	38	10	735.769	1.141	37.694
44	55	2	-0.003	1.185	0.45
54	8	2	-0.002	1.166	0.456
53	39	60	3147.566	1.142	50.347
57	5	72	941.89	1.127	53.547
63	12	57	2071.452	1.137	48.028
79	32	15	552.177	1.14	37.89
110	39	7	573.175	1.145	34.487
117	4	70	1182.809	1.138	51.532
138	11	47	785.078	1.137	48.452
139	59	43	4782.496	1.143	47.198
149	15	45	1439.679	1.141	46.299
151	17	28	889.613	1.137	45.44
153	15	21	1055.204	1.138	38.695
167	37	27	2035.956	1.141	41.247
169	8	57	1146.632	1.136	46.694
172	10	42	758.193	1.134	45.341
174	-2	41	408.982	1.125	47.509
184	31	1	-0.001	1.148	0.451
186	32	17	440.521	1.144	38.157
188	19	60	1391.888	1.139	49.213
216	2	1	0	1.163	0.453
219	7	21	1415.074	1.134	39.249
225	8	3	-0.001	1.174	0.457

4.3

Degree of control over dissemination

The degree of propagation control of a network is measured by the number of structural holes present in that network. As the network intermediate central potential data indicates the existence of partial nodes in the network located in the structural hole locations. The data of some nodes are shown inTable 2, by analyzing the effective size, efficiency coefficient, limiting system coefficient, and the degree of hierarchy, it can be found that the individuals 17, 28, 34, 42, 54, 57, 66, 108, 115, 139, 145, 166, 168, 172, 176, 190, 215, and 219 are located in the structural hole position of the network, and they are the intermediates of the network, The degree of reciprocity is high. Individuals at such nodes can also connect to two regions of the network and access non-redundant information in the network with high access.

Table 2.

Network opinion leader structure hole analysis

17	46.047	0.84	0.074	0.084
28	53.354	0.846	0.068	0.114
34	41.913	0.855	0.086	0.116
42	42.488	0.819	0.077	0.056
54	56.297	0.815	0.064	0.079
57	63.758	0.874	0.059	0.108
66	55.359	0.878	0.064	0.089
108	32.498	0.812	0.098	0.069
115	60.213	0.885	0.059	0.097
139	56.614	0.821	0.062	0.069
145	37.105	0.825	0.089	0.118
166	41.555	0.829	0.08	0.064
168	50.587	0.887	0.076	0.16
172	40.458	0.827	0.086	0.092
176	34.011	0.81	0.093	0.056
190	53.275	0.832	0.07	0.11
215	26.833	0.892	0.113	0.103
219	26.047	0.892	0.112	0.112

5

Analysis of Internet Users’ Communication Behavior of Ice and Snow Sports and Cultural Elements

5.1

Selection and calibration of variables

Before applying fsQCA for group analysis, antecedent variables need to be selected and calibrated to improve the interpretability of the findings. Combined with the previous analysis, it can be seen that in this study, self-efficacy (SE), social outcome expectancy (SOE), pleasure (PE), emotional identity (EI), sense of belonging (RE), and perceived usefulness (PU) with willingness to disseminate cultural information (IDW) were selected as antecedent variables, and the dissemination behaviors (IDB) were used as outcome variables. Second, calibration of variables refers to the transformation of raw scale data into pooled data on a scale between 0 and 1. For this study, since the scale data were all in the form of a 5-point Likert scale, the process of averaging the values of the measurement items of the antecedent variables was adopted. In the calibration of the antecedent variables, three anchor points need to be determined, namely: complete affiliation point, complete non-affiliation point and intersection point, the anchor point data are determined with reference to the 5% quartile, 95% quartile, and 50% quartile of the intersection point proposed by Ragin, and the quartile points of the calibration of the variables are as shown in Table 3, of which the 5% quartile of the IDB is the smallest of 2.45. Finally, the “Compute” component “Calibrate (x, n1, n2, n3)” function of the fsQCA3.0 software was used for the calibration of variables, and then the calibration of the calibrated variables was carried out. Finally, fsQCA3.0 software “Compute” component “Calibrate (x, n1, n2, n3)” function was used to calibrate the variables, and then the calibrated values of the study variables were centralized using the “+0.001” method. The values of the calibrated study variables were then centralized using the “+0.001” method to prevent missing case data values.

Table 3.

Variable calibration score

Predependent variable	SE	SOE	PE	EI	RE	PU	IDW	IDB
95% quantile	5.00	5.00	5.00	5.00	5.00	5.00	5.00	5.00
50% quantile	4.00	4.00	4.00	4.00	4.00	4.00	4.00	3.73
5% quantile	2.65	3.00	2.65	3.00	3.00	3.00	2.65	2.45

5.2

Truth table construction

A truth table is constructed to analyze the adequacy of the relationship between the condition variable grouping and the outcome, and the truth table can be calculated for all possible groupings, with each row referring to a possible combination. In general, the construction of the truth table can be divided into two steps. First, the initial truth table is formed based on the calibration results of the aforementioned fuzzy sets; second, the original consistency is refined based on the setting of the consistency threshold and the frequency threshold, so as to obtain the final truth table.

In the setting of the original consistency threshold, scholars at home and abroad have proposed different threshold standards for the minimum consistency, which are mainly categorized into the following three, i.e., 0.75, 0.8, and 0.85. Generally speaking, the threshold can be appropriately adjusted upward when the number of sample cases is small, and it can be appropriately adjusted downward when the number of sample cases is large. For this study, the sample size exceeds the 100 sample size of the traditional QCA study, and the consistency threshold is set to 0.8 in conjunction with the actual study. For the PRI-consistency threshold, generally speaking, it should be more than 0.5 in order to better avoid the simultaneous subset relationship, and in this study, the PRI consistency threshold was set to 0.7. In addition, the frequency of acceptable cases was set to 2, so as to realize the coding of samples meeting the above criteria as “1” and samples not meeting the above criteria as “0”. Finally, the truth table of users’ cultural information dissemination willingness/behavior that meets the conditions after the consistency threshold is obtained, and the truth table of users’ cultural information dissemination willingness/behavior is shown in Table 4.

Table 4.

User culture elements propagate behavior

SE	SOE	PE	EI	RE	PU	IDW	Number	IDB	Raw	PRI
0	1	1	1	1	1	1	81	1	0.97889	0.96871
1	1	1	1	0	1	1	6	1	0.98035	0.89407
0	1	1	0	1	1	1	8	1	0.97864	0.88945
0	0	1	1	1	1	1	2	1	0.96754	0.7882
1	1	1	1	0	1	1	2	1	0.97151	0.78534
0	1	1	0	0	1	1	2	1	0.96961	0.75707
0	1	0	1	1	1	1	4	1	0.96443	0.73196
1	0	1	1	1	1	1	1	1	0.95976	0.72047

5.3

Configuration Result Analysis

After constructing the truth table, it is possible to proceed with standardization analysis to determine the corresponding antecedent condition grouping based on the analysis results. fsQCA standardization analysis will eventually give three solutions, namely: complex solution, intermediate solution and simple solution. Most of the existing studies at home and abroad use the intermediate solution to interpret the results, so this study also takes the intermediate solution as the standard for interpreting the results, and at the same time combines the parsimonious solution with the judgment of whether an antecedent variable is the core condition of the outcome variable. Those antecedent variables that appear in both the parsimonious and intermediate solutions are generally considered to be core conditions, while variables that appear only in the intermediate solution and not in the parsimonious solution are considered to be auxiliary conditions.

The results of the intermediate and complex solutions of the standardized analysis are shown in Tables 5 and 6. The “*” in the table refers to the connection between the antecedents. “~” denotes logical non, meaning that the variable does not appear. Among them, Raw coverage refers to the original coverage, which indicates the proportion of samples in which the conditional grouping appears to the total samples, and Unique coverage refers to the unique coverage, which indicates the proportion of samples in which only samples that can be interpreted by the current conditional grouping to the total samples, and the larger the value of Unique coverage indicates that the grouping path is more likely to lead to the results.

Table 5.

Intermediate solution results

Frequency cutoff:2
Consistency cutoff:0.959805
	Raw coverage	Unique coverage	Consistency
~SEPEEIPUIDW	0.318853	0.0343125	0.960207
SOEEIREPUIDW	0.656437	0.384211	0.969875
~SESOEPE~REPU*IDW	0.248606	0.00627889	0.969376
Solution coverage:0.709325
Solution consistency:0.961517

Table 6.

Simple solution

Frequency cutoff:2
Consistency cutoff:0.959766
	Raw coverage	Unique coverage	Consistency
PEEIIDW	0.685389	0.0485678	0.960036
SOEREPU*IDW	0.673296	0.0127385	0.967285
SOEPE~EI	0.299208	0.00538869	0.918469
SOE~EIPU	0.329986	0.0120469	0.911687
SOE~EIIDW	0.327433	0.00863246	0.949568
SOEPEIDW	0.659246	0.00109605	0.970411
Solution coverag:0.795985
Solution consistency:0.920563

Based on the results obtained from the intermediate solution and the parsimonious construction presented in the table above, it is possible to clearly identify the core and auxiliary conditions in the combination of antecedent conditions. For the way the results of the group analysis are presented, they can be expressed in the form of dots instead of values. Specifically, the symbol * is used to indicate that the condition occurs only in the intermediate solution, and the symbol ¤ indicates that the condition occurs in both the intermediate and parsimonious solutions. The symbol § indicates that the condition does not exist only in the intermediate solution, and the symbol - indicates that the condition does not exist in both the intermediate and parsimonious solutions. In addition, a blank cell in the table indicates an “irrelevant” state, i.e., the existence or non-existence of the conditional variable has no effect on the outcome variable. In accordance with the above criteria, the grouping of users’ high culture information dissemination behaviors is finally integrated and summarized, and the specific grouping results are shown in Table 7.

Table 7.

The high-level information propagation behavior group state

Conditional variable	The high-level information propagation behavior group state
	C1	C2	C3
SE	§		§
SOE		¤	¤
PE	¤		¤
EI	¤	¤
RE		¤	§
PU	¤	¤
IDW	¤	¤
Consistency	0.960207	0.969875	0.969376
Raw coverage	0.318853	0.656437	0.248606
Unique coverage	0.0343125	0.384211	0.00627889
Solution coverage	0.709325
Solution consistency	0.961517

This study obtained three types of grouping patterns that contribute to users’ high cultural information dissemination behaviors, namely, grouping pattern C1: self-efficacy-pleasure-emotional identity-perceived usefulness-information dissemination willingness. Group C2: social outcome expectations - emotional identity - sense of belonging - perceived usefulness - willingness to disseminate information. Configuration C3: self-efficacy-social outcome expectations-pleasure-sense of belonging-perceived usefulness-willingness to disseminate information. From the results of the constructs, it can be seen that in each type of grouping pattern, high perceived usefulness and high willingness to disseminate cultural information are the core conditions triggering users to generate information dissemination behaviors of ice and snow sports and cultural elements, indicating that when users browse information related to ice and snow sports and cultural elements, the perceived usefulness and the subsequent willingness to disseminate the information in the content can highly contribute to the emergence of their dissemination behaviors. This illustrates the importance of the two types of grouping in the three types of grouping, and to a certain extent, corroborates the results of the quantitative analysis of the aforementioned structural equation modeling.

In terms of the coverage of the three types of groupings, the coverage of grouping mode 2 is 0.656437, which is much higher than that of grouping mode 1 (0.318853) as well as grouping mode 3 (0.248606), so this study concludes that the core conditions of grouping mode 2, namely, social outcome expectations, emotional identity, sense of belonging, as well as perceived usefulness and willingness to disseminate cultural information, are important factors that motivate the users to produce high cultural information dissemination behavior. In addition, except for the sense of belonging, social outcome expectation and sense of belonging also exist as core conditions in both types of grouping patterns, so it can be considered that in the grouping pattern of high cultural information dissemination behavior, social outcome expectation, emotional identity and perceived usefulness should be the three core conditions that relevant parties need to pay attention to, and the user’s cultural information dissemination behavior in social situations is more affected by the combined effect of these antecedent conditions. Users’ cultural information dissemination behaviors in social contexts are more affected by the combination of these antecedent conditions.

6

Conclusion

This paper introduces potential propagation nodes on the basis of the SIR model, improves the SIR model, and constructs the propagation method of social network of ice and snow sports and cultural elements based on the SPIR model. 1)

Ice and snow sports and cultural elements of the theme community network opinion leaders have differences in information dissemination ability. There are 8 social leaders in the network opinion leaders, their numbers are 14, 26, 53, 63, 139, 167, 188, 219, and they exist in the core area of the network. The information dissemination ability of this part of individuals is better than those in the edge region. 18 individuals are located in the structural hole position of the network and are the intermediaries of this network with a high degree of reciprocity. It shows the high degree of communication control of this social network, which has the function of regulating and controlling the information of the network.

2)

Using the fsQCA tool and based on the theory of group state analysis, three types of conditional grouping for users’ high cultural information dissemination behaviors are derived, and the coverage of group state mode 2 is 0.656437, which is much higher than that of group state mode 1 and group state mode 3. Therefore, it can be assumed that the core conditions contributing to the three types of grouping of high cultural information dissemination behaviors related to the elements of ice, snow and sports culture are Social outcome expectations, emotional identity and perceived usefulness.

Sprache:: Englisch

Zeitrahmen der Veröffentlichung:: 1 Hefte pro Jahr
Fachgebiete der Zeitschrift:: Biologie, Biologie, andere, Mathematik, Angewandte Mathematik, Mathematik, Allgemeines, Physik, Physik, andere

Zeitschrift RSS Feed

Research on the Intelligent Optimization Method of Digital Communication Path of Ice and Snow Sports and Cultural Elements

Liang Han

Online veröffentlicht: 19. März 2025

Eingereicht: 27. Okt. 2024

Akzeptiert: 13. Feb. 2025

DOI: https://doi.org/10.2478/amns-2025-0477

SchlüsselwörterSocial Networks, Information Dissemination, Infectious Disease Modeling, SPIR Modeling, Ice Sports Culture

© 2025 Liang Han, published by Sciendo

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

Schlüsselwörter
Social Networks, Information Dissemination, Infectious Disease Modeling, SPIR Modeling, Ice Sports Culture