Information propagation analysis method based on three-dimensional network information propagation dynamics model

Abstract

The application provides an information propagation analysis method and system based on a three-dimensional network information propagation dynamics model, which characterizes multiple factors affecting information propagation as a social network layer, an environment layer, a psychological layer and a viewpoint layer in a supernetwork, constructs a coupling network for the interaction phenomenon between users in the social network layer, thereby generating a three-dimensional network of supernetwork nested coupling networks, and constructs an information propagation dynamics model by fusing rich propagation mechanisms on the basis of the network. Real information propagation case events are used to carry out parameter estimation, an information propagation index system is constructed, key factors affecting information propagation evolution are analyzed, and finally a theoretical framework is provided for accurately grasping information propagation rules in the field of network information propagation analysis, and technical and strategic support is provided for network information propagation response and guidance.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information dissemination, more specifically, to an information dissemination analysis method and system based on a multi-element three-dimensional network information dissemination dynamics model of a real social network. BACKGROUND

[0002] In recent years, under the rapid development of mobile internet technology, various online social media have developed rapidly. With the help of these platforms, the public can express their views relatively freely through text, and at the same time, the barriers of various network social platforms are gradually being eliminated, and netizens can access information content on multiple social platforms while expressing their own opinions, and can also perform one-way or two-way cross-platform interaction through "forwarding", "sharing" and other methods, which greatly reduces the threshold for ordinary people to express their opinions and express their opinions through the Internet, making public opinion expression more smooth.

[0003] In addition, as an external driving force for the process of information dissemination, the information dissemination environment is also more complex. In addition to the online and offline environment as an external force to drive the dissemination of network information, the influence of offline information environment created by radio, television and other media and activities cannot be ignored. Corresponding to the external driving force, the internal driving force focuses on the personal characteristics of users, and the psychological characteristics and views of users will play an internal driving role in the dissemination of network information. Due to the differences in social relations and personal cognition, netizens have different personal characteristics in information dissemination events, which has an increasingly significant impact on their information dissemination decisions. On the other hand, the expression of netizens is also more diversified, and user-generated content is more personalized. The expression of netizens is no longer limited to neat literary language, and users will create popular language, emoticons and other rich content to further promote the sustainable dissemination of network information.

[0004] From the overall evolution trend, the dissemination of network information has dynamic characteristics, and the dissemination mechanism is very important in forming the evolution mode of public opinion. The elements affecting the process of network information dissemination are also diversified from the two angles of internal and external driving forces. Network information dissemination dynamics theory takes dynamic systems as the research object and can well describe the dissemination mechanism, influencing factors, dissemination rules, etc. Therefore, exploring the information dissemination rules based on this theory has become a hot research direction.

[0005] However, existing research often focuses on a single perspective rather than comprehensively considering the combined impact of multiple factors on information dissemination, resulting in unclear network structures and simplistic considerations of dissemination mechanisms. Furthermore, most studies aim to build specific models based on research objectives, failing to conduct multi-scenario information dissemination research within a general framework. This leads to incomplete multi-scenario simulations that cannot meet the needs of understanding the laws governing information dissemination across various scenarios. Finally, current research often underutilizes real-world data and lacks an efficient and accurate parameter estimation method, resulting in an inaccurate grasp of the laws governing information dissemination in real-world scenarios and an inability to reproduce real-world information dissemination scenarios using models. Summary of the Invention

[0006] To address the aforementioned problems in existing network information dissemination analysis, this invention, based on the aforementioned information dissemination scenarios and technologies, combines theories such as information dissemination dynamics and complex networks to construct a multi-element, three-dimensional network information dissemination dynamics model. It performs numerical fitting and analysis of the model based on real information dissemination data, studies the impact of model parameters on model indicators through parameter change analysis techniques, summarizes the laws of information dissemination, and thus formulates information dissemination guidance and control strategies for research within a general framework.

[0007] According to one aspect of the present invention, an information propagation analysis method based on a three-dimensional network information propagation dynamics model is provided. This method is used to simulate the state transitions of individual netizens and the evolutionary trends of information propagation in network information propagation events based on a preset three-dimensional network information propagation dynamics model.

[0008] The structure of the three-dimensional network includes:

[0009] The social network layer is used to represent the directed complex network structure of netizens participating in the information dissemination process through a set of a finite number of information dissemination subject nodes and the directed edge relationships between nodes. Here, netizens on each social platform are nodes, the edges between nodes represent the dissemination of information between users in the three-dimensional network information dissemination, the edges pointing to user nodes represent the influence of other users on the user, and the edges radiating outward from the user node represent the user's ability to disseminate information outward.

[0010] The environment layer is used to represent the information dissemination environment of the beginning, development, climax and ending of the information dissemination event through a set of finite environment type nodes and directed edge relationships between nodes; wherein, the directed edges of the environment layer represent the temporal relationship of environmental information release, and the directed connections between multiple environment nodes represent the continuous changes in the environment;

[0011] a psychological layer for representing the psychology of the netizen in the information dissemination event and the transformation relationship between the psychological types through a set of a limited number of psychological type nodes and directed edges between the nodes, the psychological state transition of the psychological layer being affected by the environment layer;

[0012] a viewpoint layer for representing the transformation of the viewpoint of the netizen in different periods of the information dissemination event through a set of a limited number of viewpoint type nodes and directed edges between the nodes; the viewpoint of the viewpoint layer being divided into positive and negative viewpoints according to the viewpoint theme and emotional tendency;

[0013] the super edge SE of the three-dimensional network represents the subject a i under the joint action of the external driving force e m and the internal driving force p n of the psychological state, the viewpoint k j is published, and finally e m , p n , k j influence the change of the dissemination state of the information dissemination subject a i in the social network layer; the three-dimensional network information dissemination dynamics model takes the time sequence relationship of the development process of the event in the environment layer as the main line, takes the other factors in the environment layer as a whole as the external driving force, takes the nested influence of the psychological layer and the viewpoint layer as the internal driving force, and is based on the information dissemination general scenario formed under the joint action of the internal and external driving forces to influence the dissemination state evolution process of the user node in the social network layer;

[0014] The method comprises:

[0015] determining the key time points in the development process of the information dissemination event based on the selected information dissemination event;

[0016] extracting the hot topic event information of the information dissemination event at the key time points and the forwarding information in the forwarding area of the hot topic event information as original dissemination data; wherein the forwarding information includes the forwarding text and forwarding time of the forwarding user under the hot topic event information;

[0017] preprocessing the original dissemination data to obtain the dissemination effective data in the original dissemination data;

[0018] based on the dissemination effective data and the preset three-dimensional network information dissemination dynamics model, simulating the group state transition and information dissemination event development evolution dynamic process under the three-dimensional network of the social media platform.

[0019] In another aspect, the present application also provides an information propagation analysis system based on a three-dimensional network information propagation dynamics model, which is used for information propagation analysis by using the information propagation analysis method based on a three-dimensional network information propagation dynamics model as described above, comprising:

[0020] An original propagation data acquisition unit is configured to determine a key time point in the development process of the selected information propagation event, and extract hot topic event information of the information propagation event at the key time point and forwarding information in a forwarding area of the hot topic event information as original propagation data, wherein the forwarding information comprises forwarding text and forwarding time of a forwarding user under the hot topic event information.

[0021] A data preprocessing unit is configured to preprocess the original propagation data to obtain propagation effective data in the original propagation data.

[0022] An analog analysis unit is configured to simulate group state transition and information propagation event development evolution dynamic process under a three-dimensional network of the social media platform based on the propagation effective data and a preset three-dimensional network information propagation dynamics model.

[0023] The information propagation analysis method and system based on a three-dimensional network information propagation dynamics model according to the present application described above characterize multiple factors affecting information propagation as a social network layer, an environment layer, a psychological layer and a viewpoint layer in a super network, construct a coupling network for the interaction phenomenon between users in the social network layer, generate a three-dimensional network of a super network nested coupling network, and construct an information propagation dynamics model by fusing rich propagation mechanisms on the basis of the network. Real information propagation case events are used to carry out parameter estimation, an information propagation index system is constructed, key factors affecting information propagation evolution are analyzed, and finally a theoretical framework for accurately grasping information propagation rules in the field of network information propagation analysis is provided, and technical and strategic support for network information propagation response and guidance is provided.

[0024] To the accomplishment of the foregoing and related ends, one or more aspects of the application, as generally described herein, include the features recited in the claims below and as detailed below in the detailed description. These and other aspects of the application will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, illustrating by way of example the principles of the application. It is to be understood that this application is not limited to the methodology set forth and / or suggested herein, but that the application is applicable to any medication delivery regimen employing principles of the application. BRIEF DESCRIPTION OF DRAWINGS

[0025] Other objects and results of the application will become more fully understood and appreciated with reference to the following detailed description, taken in conjunction with the accompanying drawings, in which:

[0026] Figure 1 A network information propagation schematic diagram according to an embodiment of the present application;

[0027] Figure 2 A method flow schematic diagram for information propagation analysis based on a three-dimensional network information propagation dynamics model according to an embodiment of the present application;

[0028] Figure 3 A three-dimensional network structure schematic diagram of the MF-SLPCI model as a whole according to an embodiment of the present application;

[0029] Figure 4 A social network layer node propagation state transition process schematic diagram of the MF-SLPCI model according to an embodiment of the present application;

[0030] Figure 5 A propagation accumulation and propagation effective amount curve schematic diagram according to an embodiment of the present application;

[0031] Figure 6 A case event information propagation cycle schematic diagram according to an embodiment of the present application;

[0032] Figure 7 A selected information microblog forwarding text view distribution situation schematic diagram according to an embodiment of the present application;

[0033] Figure 8 A numerical fitting result schematic diagram of four selected information according to an embodiment of the present application;

[0034] Figure 9 An environmental impact related parameter analysis schematic diagram according to an embodiment of the present application;

[0035] Figure 10 An individual characteristic related parameter analysis schematic diagram according to an embodiment of the present application;

[0036] Figure 11 A network structure related parameter analysis schematic diagram according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more embodiments. It can be evident, however, that embodiments can be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate describing one or more embodiments.

[0038] In view of the above problems existing in the existing network information propagation analysis, in view of the above problems existing in the existing network information propagation analysis, the present application is based on the kinetic theory and the diversified information propagation scene in the real world, taking the network information propagation dynamics theory as the cornerstone, based on the internal and external driving force elements such as social network user interaction, user psychology, user viewpoint and information propagation environment, and combining the influence of various real factors on information propagation, a multi-element based three-dimensional network information propagation dynamics theory is proposed.

[0039] The multi-element based three-dimensional network information propagation dynamics theory provided by the present application can study the internal driving force and external driving force affecting information propagation in layers, further combine external driving force and internal driving force elements to construct a three-dimensional network including social network layer, environment layer, psychological layer and viewpoint layer, and further set the propagation mechanism based on individual factors, network structure factors and the like to construct a multi-element based three-dimensional network information propagation dynamics model general framework to describe the information propagation process in a complex system. With real information propagation cases as data driving, the crowd time series cumulative amount is obtained by using statistical analysis method and the real data is fitted to verify the effectiveness of the model, then the influence of various physical factors on information propagation is further considered from multiple angles, and the information propagation law is summarized, thereby providing decision support for related information guiding strategy.

[0040] The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0041] The network information propagation under the new propagation pattern can be summarized as: the sum of the viewpoints published by the information propagation subject on the social platform under the influence of the internal driving force of psychology and the external driving force of environment. Figure 1 For the network information propagation schematic diagram, as shown in Figure 1 The network information is rapidly developed under the influence of various elements such as propagation carrier, external driving force, internal driving force and performance mode. At present, the research on the influence of various elements on information propagation is very extensive, but most of the research does not comprehensively consider various elements to form a general framework suitable for rich and multi-scene information propagation cases.

[0042] The present application is based on internal and external driving forces and real data to drive netizens to participate in information dissemination development, and constructs a multi-factor based three-dimensional network information dissemination dynamics model to analyze the information dissemination law under the influence of multi-factors. Through the study of the multi-factor based three-dimensional network information dissemination dynamics model MF-SLPCI (Muti-factors Susceptible-Latent-Promoting-Countering-Immune) (hereinafter also referred to as MF-SLPCI model or multi-factor based three-dimensional network information dissemination dynamics model), a general framework capable of accurately simulating the state transition of netizen individuals and the evolution trend of information dissemination in real network information dissemination events is built.

[0043] Figure 3 The schematic diagram of the overall three-dimensional network structure of the MF-SLPCI model according to the embodiment of the present application. With the gradual elimination of barriers between social platforms, netizens can obtain information content on multiple social platforms while expressing their own opinions, and can also perform one-way or two-way cross-platform interaction through "forwarding", "sharing", "cooperation" and the like. At the same time, netizens are not only affected by the social platform environment and friends, but also affected by the external information dissemination environment created by media such as radio, television and activities. From the perspective of netizens, the psychology, opinions and other characteristics of netizens as internal forces also have a driving effect on network information dissemination. Due to the differences in circle culture, social relations and the like, netizens have different views on information events, and the influence of the psychological, opinion and other attribute characteristics of netizens on their information dissemination decisions is increasingly significant.

[0044] As shown in Figure 3 , the multi-factor information dissemination scene under the joint action of internal and external driving forces can be described as: the information dissemination subjects on the social network, the information dissemination environment information, the psychological types of netizens, and the opinions of netizens can be abstracted into nodes in the social network layer, the environment layer, the psychological layer, and the opinion layer in the super network (three-dimensional network). The connection between homogeneous nodes constitutes a subnetwork, and the connection between heterogeneous nodes constitutes the connection between subnetworks. In the three-dimensional network, the social network layer represents the dissemination state attribute of the netizen, the environment layer represents the information dissemination environment attribute in which the netizen is located, the psychological layer represents the psychological state attribute of the netizen, and the opinion layer represents the opinion expression attribute of the netizen.

[0045] Specifically, as an example, in the MF-SLPCI model, the specific structure of each layer is as follows:

[0046] (1) Social network layer

[0047] The social network layer can be regarded as a directed complex network structure composed of netizen nodes participating in the information dissemination process, and can be represented as where A n= {a1, a2,..., a m} represents a set of a finite number of information dissemination subject nodes, The directed edge relationship between nodes in the social network layer. In the social network layer, nodes represent netizen individuals on each platform, and the edges between nodes represent the participation of netizens in three-dimensional network information dissemination. The propagation of information between users is regarded as an edge, and the edge pointing to the user node represents the influence of other users on the user, and the edge diverging from the user node represents the user's ability to disseminate information outward.

[0048] (2) Environment layer

[0049] The node types in the environment layer comprehensively represent the information dissemination environment at the beginning, development, climax and end of the event, and can be represented as S E = (E, E E→E ), where E = {e1, e2,..., e n} represents a set of a finite number of environment type nodes, and E E→E represents the directed edge relationship between nodes. The directed edge in the environment layer represents the time sequence relationship of environmental information release, and the directed edge between multiple environment nodes represents the continuous change of the environment.

[0050] (3) Psychological layer

[0051] The psychological layer can be represented as S P = (P, E P→P ), where P = {p1, p2,..., p n} is a set of a finite number of psychological type nodes, and E P→P is a directed edge relationship between nodes. The directed edge between nodes in this layer represents the transformation relationship between each psychological type, and p i The psychological transformation into p j has a certain probability value, and this value is affected by other layers. Each node type represents the netizen's "conformity", "reservation", and "opposition" psychology towards information dissemination events, and the psychological state transition of this psychological layer is also affected by the environment layer.

[0052] (4) View layer

[0053] The view layer can be represented as S K = (K, E K→K ), where K = {k1, k2,..., k n} is a set of a finite number of view type nodes, and E K→Kis a directed edge between nodes. In the process of information dissemination, the public can express their views and opinions through text, express their emotions, and use directed edges to represent the transformation of the views held by netizens at different stages of information dissemination events. The view type can be divided according to the view theme and emotional tendency: type a represents a positive emotional tendency, indicating that the netizen supports the view; type b represents a negative emotional tendency, indicating that the netizen opposes the view.

[0054] Overall, the super-edge of the stereoscopic network contains four types of heterogeneous nodes, and the super-edge SE represents the subject a i In the information dissemination environment, this external driving force e m and the internal driving force p n of the psychological state jointly act on the published view k j , and finally e m , p n , k j influence the change of the dissemination state of the information dissemination subject a i in the social network layer, which can be represented by formula (1):

[0055] SE={a i , D{e m , p n}, k j | θ(a i , k j )=1, θ(a i , p n )=1, θ(a i , e m )=1} (1)

[0056] Based on the above analysis, the stereoscopic network of the embodiment can be represented as F={S A , S E , S P , S K , SE}. Based on the above-mentioned constructed stereoscopic network, the embodiment first takes the time sequence relationship of the development process of the environment layer event as the main line, and considers the other factors of the environment layer as a whole as an external driving force, and the nested influence of the psychological layer and the view layer as an internal driving force. Under the joint action of the internal and external driving forces, the dissemination state evolution process of the user nodes in the social network layer is affected, and then the MF-SLPCI model is constructed based on the general information dissemination scenario.

[0057] Figure 2 The flow of the information dissemination analysis method based on the stereoscopic network information dissemination dynamics model according to the embodiment of the application is shown. As Figure 2As shown, on the basis of the above-mentioned three-dimensional network information propagation dynamics model, the information propagation analysis method based on the three-dimensional network information propagation dynamics model provided by the application simulates the state transition of netizen individuals and the evolution trend of information propagation in the network information propagation event based on the pre-constructed three-dimensional network information propagation dynamics model, and specifically includes the following steps:

[0058] S210: determining a key time point in the development process of the information propagation event based on the selected information propagation event;

[0059] S220: extracting hot topic event information of the information propagation event at the key time point and forwarding information in the forwarding area of the hot topic event information as original propagation data; wherein the forwarding information includes the forwarding text and forwarding time of the forwarding user under the hot topic event information; the hot topic event information can be the first information published by the topic host under the hot topic;

[0060] S230: preprocessing the original propagation data to obtain propagation effective data in the original propagation data;

[0061] S240: simulating the state transition of the group under the three-dimensional network of the social media platform and the development evolution dynamic process of the information propagation event based on the propagation effective data and the pre-set three-dimensional network information propagation dynamics model.

[0062] The pre-constructed three-dimensional network information propagation dynamics model will be described in more detail in the following more specific implementation mode.

[0063] In the social network layer, the microblog users participating in the information propagation process are taken as nodes, and according to the role of individuals in the information propagation, the nodes can be mainly divided into the following six categories: susceptible S, latent L, supporter P, opponent C and immune I. Under the action of factors such as three-dimensional network structure, influence factor, propagation mechanism and interaction rules between individuals, the state of the nodes in the social network layer will change. In the topology structure of the social network layer, different types of nodes play different roles in the process of information propagation, and the social relationship between the nodes provides a channel and a way for them to carry out interaction, such as Figure 4 As shown.

[0064] According to the node propagation state transition process of the MF-SLPCI model social network layer shown in Figure 4 As shown, the multi-factor three-dimensional network state conversion process can be described as follows:

[0065] Assuming that the information is spread in a closed stable environment, only considering the population that can be reached in the process of information spreading itself. This embodiment focuses on the information diffusion generated by the individual forwarding behavior of each piece of information, assuming that the same user can choose whether to forward the information after reading it, and a user forwards each piece of information only once. At any moment, any individual in the population can be in six states with respect to the information, namely, susceptible (abbreviated as S), latent (abbreviated as L), promoting (abbreviated as P), countering (abbreviated as C), and immune (abbreviated as I). This embodiment introduces the in-degree k of the cubic network node, that is, the number of edges pointing to the node in the cubic network, and the out-degree l of the cubic network node, that is, the number of edges emitted by each node in the cubic network, and classifies the population in each state according to the in-degree and out-degree of different nodes in the cubic network, wherein,

[0066] S (k，l) (t) represents the total amount of individuals in the susceptible state with in-degree k and out-degree l at time t, which has not yet been exposed to the information but has the opportunity to be exposed to the information in the future and is susceptible to the information, and may have a forwarding behavior;

[0067] L (k，l) (t) represents the total amount of individuals in the latent state with in-degree k and out-degree l at time t, which has been exposed to the information but has not yet spread the information;

[0068] P (k，l) (t) represents the total amount of individuals in the promoting state with in-degree k and out-degree l at time t, which has forwarded the information and has a consistent view with the original information, and is still within the exposure period of the information, and has the ability to make individuals in the susceptible state aware of the content of the information and may have a forwarding behavior;

[0069] C (k，l) (t) represents the total amount of individuals in the countering state with in-degree k and out-degree l at time t, which has forwarded the information and has an inconsistent view with the original information, and is still within the exposure period of the information, and has the ability to make individuals in the susceptible state aware of the content of the information and may have a forwarding behavior;

[0070] I (k，l)(t) represents the total number of individuals in the immune state of the information with in-degree k and out-degree l at time t. This population consists of two parts: one part is the population that has forwarded the information and exceeds the exposure period of the information over time, and the total number of individuals that no longer have the ability to affect other users to contact the content of the information and produce forwarding behavior; the other part is the total number of individuals in the susceptible state, latent state, support forwarding state and opposition forwarding state that contact the support forwarding state individuals and opposition forwarding state individuals, and directly transform into the total number of individuals immune to the information due to subjective disinterest in the information.

[0071] Therefore S ( k, l ) (t), L ( k, l ) (t), P ( k, l ) (t), C ( k, l ) (t), I ( k, l ) (t) respectively represent the number of each population with in-degree k and out-degree l at time t, and the sum of the total number of each population at time t is N(t), N(t) is a function of time.

[0072] The MF-SLPCI model constructed in this embodiment is:

[0073]

[0074] p = p P + p C (7)

[0075]

[0076] The initial values of each state are defined as: S (k，l) (0) = S0, L (k，l) (0) = L0, P (k，l) (0) = C P (0), C (k，l) (0) = C C (0), I (k，l) (0) = 0.

[0077] Wherein, m ((i, j) | (k, l)) represents the conditional probability of any edge in the directed network being emitted by the node with "degree" (i, j) and pointing to the node with "degree" (k, l), (i.e. p P ) represents the probability of the node user with "degree" (k, l) being connected with the user in the support forwarding state at time t, (i.e. p C) represents the probability that the node user of degree (k, l) is connected to the user in the opposite forwarding state at time t, and p represents the probability of being connected to the forwarding state user. For a multi-layered three-dimensional network, the directed network is a certain layer in the three-dimensional network, and the above specific state transition formula is for the user node relationship of the social network layer in the three-dimensional network. In this embodiment, it is assumed that the directed network is an irrelevant network, and the conditional probability is only related to the degree of the node (i, j) with the upstream in-degree i and the out-degree j, which can be written as:

[0078] m((i, j) | (k, l)) = j m(i, j) / p <l>(10)

[0079] where (i, j) represents a user node with different in-degree i and out-degree j, respectively corresponding to a node with in-degree i and out-degree j; m(i, j) represents the joint probability distribution of node (i, j) in the network, that is, the probability of randomly selecting a node with in-degree i and out-degree j in the network, which is also equal to the total number N of nodes with in-degree k and out-degree l in the network kl The ratio of the total number N of network nodes, <l>The average degree of the directed network is denoted by k; the degree distribution of any point in the network is (k, l).

[0080]

[0081] Because in any directed network, each edge is from one node to another, the average in-degree is equal to the average out-degree in the directed network, which is expressed as:

[0082]

[0083] which can be simplified as:

[0084]

[0085] which respectively represent the probability of a node user with a degree of (k, l) connecting with a user supporting forwarding and a user opposing forwarding in a unit of time. The remaining parameters in the formula are explained in Table 1.

[0086] Table 1. Parameter explanation of the three-dimensional network model

[0087]

[0088]

[0089] In this embodiment, it is assumed that the users newly migrated to the dynamic system are all susceptible, and therefore, the average migration rate Λ is added to the change rate of the number of susceptible persons reflected in formula (1). In addition, it is assumed in this section that each type of group leaves the dynamic system at an average migration rate u. After the susceptible persons S and the hesitant persons L are influenced by the supporters P and the opponents C, there will be four state transition paths, which are to become hesitant persons L, supporters P, opponents C, and immune persons I. After the supporters P and the opponents C are influenced by the supporters P and the opponents C, there will be three state transition paths, which are to become supporters P, opponents C, and immune persons I. In the process of the state transition of the netizens, in addition to the influence of the social network layer, the environment layer, the psychological layer, and the viewpoint layer will also determine the transmission state of the netizens. The specific three-dimensional network parameter settings are shown in formulas (14)-(24), and the related parameters involved are shown in Table 2, and the function setting mechanism is shown in Table 3.

[0090] f S→L (t) = τ × (1 - e -δ×n ) (15)

[0091]

[0092] f P→I (t) = 1 - f P→C (t) (22)

[0093] f C→I (t) = 1 - f C→P (t) (23)

[0094] f S→I (t) = 1 - f S→L (t) - f S→P (t) - f S→C (t) (24)

[0095] f L→I (t) = 1 - f L→P (t) - f L→C (t) (25)

[0096] Table 2 Definition of parameters of the stereoscopic network

[0097]

[0098] Table 3 Definition of time-varying parameters of the stereoscopic network

[0099]

[0100]

[0101] Specifically, in the dynamic process of the stereoscopic network propagation, an average user in a susceptible state will be affected by k individuals per unit time. Since in the degree-independent directed network, the conditional probability of a user with degree (k, l) being connected to another supporting / opposing forwarding user is m((i, j) | (k, l)), therefore the probability of a user with degree (k, l) in a susceptible state contacting a user in a forwarding state per unit time is k(p P + p C ), when k pS (k，l) (t) users contact supporters / opponents, wherein:

[0102] k p f S→L (t) S (k，l) (t) susceptible individuals will maintain sustained attention to the information, i.e., from the S state to the L state;

[0103] k p f S→P (t) S (k，l) (t) susceptible individuals will support the forwarding of information with the same view as the original information, i.e., from the S state to the P state;

[0104] k p f S→C (t) S (k，l) (t) susceptible individuals will oppose the forwarding of information with a view inconsistent with the original information, i.e., from the S state to the C state;

[0105] The rest of the susceptible are not interested in the information, i.e. kpf S→I (t) S (k，l) (t) susceptible individuals transfer from S state to I state.

[0106] During the incubation period, θL (k，l) (t) latent individuals enhance the willingness to spread the information, thus generating the forwarding behavior, i.e. f L→P (t) θ(t) L (k，l) (t) latent individuals transfer from L state to P state, f L→C (t) θ(t) L (k，l) (t) latent individuals transfer from L state to C state; the rest of f L→I (t) θ(t) L (k，l) (t) latent individuals lose interest in the information, i.e. transfer from L state to I state. For the forwarders, the supporters P / opponents C, when influenced by the same supporters P / opponents C, will have kpf P→C (t) P (k，l) (t) supporters transfer from P state to C state kpf C→P (t) C (k，l) (t) opponents transfer from C state to P state, another part of the supporters P and opponents C will have kpf C→I (t) P (k，l) (t), kpf C→I (t) P (k，l) (t) the number of supporters / opponents transfer from P / C state to I state, the rest of α P (t) P (k，l) (t) / α C (t) C (k，l) (t) supporters / opponents no longer have the ability to influence others due to the passage of the active exposure period, i.e. transfer from P / C state to I state.

[0107] The forwarding cumulative amount can be directly obtained from the network transmission platform for calculation. In the three-dimensional network information transmission dynamics model, the change rate of the forwarding cumulative amount of the support forwarding users and the opposition forwarding users with degree (k, l) with time is:

[0108]

[0109] Then the forwarding cumulative amount C P (t) and C C (t) of the supporters and the opponents are respectively:

[0110]

[0111] In the process of information dissemination, how to measure the effectiveness of public information dissemination and the degree of opinion evolution is a major focus of this invention's research on the evolutionary laws of network information dissemination. In a specific embodiment of this invention, the cumulative amount of group dissemination and the effective amount of group dissemination are used to characterize the dissemination trend from two aspects: the breadth of dissemination and the intensity of dissemination, respectively. The cumulative amount of group dissemination represents the total number of people in the system who have received the information up to a certain point in time, while the effective amount of group dissemination represents the number of people in the system who have received the information and still have influence up to a certain point in time. Therefore, the curves characterizing the changes of these two quantities over time can be defined as dissemination curves. Figure 5 The following propagation information can be obtained from the support / resistance diagram of the cumulative propagation amount and effective propagation amount curves shown:

[0112] (1) Peak of information dissemination popularity

[0113] Peak Information Dissemination Popularity P max / C max It plays a crucial role in describing the popularity of information shared by a group, representing the largest group still capable of influencing others, corresponding to the peak value of the effective propagation curve P(t) / C(t). max / C max The larger the value, the higher the popularity of the information during its dissemination.

[0114] (2) The final scale of information dissemination

[0115] Ultimate scale of information dissemination It plays an important role in describing the breadth of group propagation, representing the total number of groups in state F during the entire information propagation process, and corresponding to the final value of the cumulative propagation curve P(t) / C(t). The larger the value, the wider the scope of information dissemination that supports (opposes) the forwarder during the development of information dissemination.

[0116] (3) Peak time of information dissemination

[0117] Information dissemination peak time t max This indicates the moment when the information reaches its peak popularity, satisfying the condition P(t). max ) = P max ,C(t) max ) = C max .

[0118] like Figure 5 As shown, its cumulative propagation curve C is similar to an "S"-shaped curve, specifically showing a trend of rapid rise, slow rise, and finally stabilization, while its effective propagation curve F is similar to a bell-shaped curve (normal curve), specifically showing a trend of rise first and then fall.

[0119] It can be seen from the above specific embodiment of constructing the three-dimensional network information propagation dynamics model that the present application is based on the kinetic theory and diversified information propagation scenarios in the real world, takes the network information propagation dynamics theory as the cornerstone, is based on the internal and external driving force elements such as social network user interaction, user psychology, user viewpoint and information propagation environment, fuses the influences of various real factors on information propagation, and proposes a three-dimensional network information propagation dynamics theory based on multiple elements. The theory studies the internal and external driving forces affecting information propagation in layers, further combines elements such as external driving force, internal driving force and performance mode to construct a three-dimensional network including a social network layer, an environment layer, a psychology layer and a viewpoint layer, and further sets a propagation mechanism based on individual factors and network structure factors to construct a general framework of the three-dimensional network information propagation dynamics model based on multiple elements to describe the information propagation process in a complex system. Real information propagation cases are taken as data driving, statistical analysis methods are used to obtain crowd time series accumulations and real data fitting is performed to verify the effectiveness of the model, then the influences of various physical factors on information propagation are comprehensively considered from multiple angles, information propagation rules are summarized, and decision support is provided for government related information propagation guiding strategies.

[0120] On the basis of the above pre-constructed three-dimensional network information propagation dynamics model, the state transition of network users and the evolution trend of information propagation in a network information propagation event are simulated. First, based on a selected information propagation event, key time points in the development process of the information propagation event are determined, and hot topic event information of the information propagation event at the key time points and forwarding information in the forwarding area of the hot topic event information are extracted as original propagation data. This process corresponds to steps S210 and S220.

[0121] Specifically, as an example, a selected information propagation event is analyzed through a preset information analysis website "Zhiwei event", the event heat map trend and the heat and data volume of the corresponding microblog information are combined, and the key time points in the development process of the information propagation case are selected, such as Figure 6 The red box in the figure. Through the application programming interface open to the outside world of the social microblog platform, 4 hot topic host information at the key time node in the case were collected, which were: case information 1 of the starting event topic of the information dissemination starting at 10:02 on June 9, 2023, case information 2 of the information dissemination development stage on June 9, 2023, case information 3 of the information dissemination further development stage on June 11, 2023, and case information 4 on June 11, 2023. For a specific message, the relevant data including the information content and the forwarding time of each user were collected through the API (Application Programming Interface) interface open to the outside world of the social microblog platform.

[0122] In this embodiment, the hot topic event information at the key information dissemination development time point was selected within 60 hours from the case information dissemination starting to ferment to the event heat subsiding, and the forwarding information in the forwarding area was further collected, including the forwarding text of the forwarding user under the information and the forwarding time, to provide data support for verifying the effectiveness of the model.

[0123] Then, the original propagation data is preprocessed to obtain the propagation effective data in the original propagation data, and this process corresponds to the above step S230.

[0124] Specifically, as an example, the present application carries out data preprocessing on each single information data collected at the key node in the information dissemination process, which includes three steps. First, due to physiological factors, users are mainly active in browsing information during the day, and accordingly the night data of the original data is processed to avoid the limitation of information stagnation caused by physiological needs. Second, due to the randomness of user forwarding information, the data is filtered again, and the records irrelevant and meaningless to the forwarding text information are deleted. Finally, the SnowNLP library is used to quantify the forwarding text opinion value under the four messages to realize the division of supporters and opponents in the SLPCI model. The specific analysis results are as shown in Figure 7 The information forwarders corresponding to the texts in the interval [0-0.5] are divided into opponents, and the information forwarders corresponding to the texts with the quantified opinion value results between [0.6-1] are divided into supporters. As shown in Figure 7 The results show that the information of four key nodes of information dissemination is concentrated in the interval [0.6-0.7], which is also related to the nature of the event. The information publishers at each time point are official accounts, and the information content is positive, which has a certain reference value, so more information forwarders hold a supporting attitude. After these preprocessing steps, the effective data of propagation is obtained, that is, the current forwarding time point and content of supporters and opponents in each message, which is used to calculate the cumulative forwarding quantity. In this embodiment, the start time is set to 0, and the sampling frequency is set to one hour.

[0125] Then, the group state transition and information propagation event development and evolution dynamic process under the stereoscopic network can be simulated based on the propagation effective data and the preset stereoscopic network information propagation dynamics model, which corresponds to the above step S240 and reflects the group propagation state transition and information propagation event development and evolution dynamic under the influence of the social media platform network environment and the individual factors of netizens.

[0126] In this embodiment, the cumulative forwarding quantity is first used as an information propagation index to perform a numerical fitting experiment to estimate the model parameters, and then the coincidence degree and the corresponding error index of the real data and the model fitting effect are analyzed to determine whether the model is applicable to simulate the information propagation process, and further to verify the effectiveness of the model.

[0127] In the numerical fitting experiment, the nonlinear least square (LS, Least Square) is used for parameter estimation, and the numerical method is used to seek the optimal solution of parameter estimation and data fitting. In this process, the initial value vector of the parameters that need to be estimated in the stereoscopic network information propagation dynamics model proposed in the present application is set as Θ=(τ, δ, n, η P , λ P , η C , λ C , K max , L max , Λ, u, θ, α P , α C , S0, L0, N). The LS error function is obtained as

[0128]

[0129] wherein, represents the numerical result of C P (t) under the parameter vector Θ, corresponding to formula (27), represents the real cumulative forwarding quantity of supporters; represents the numerical result of C C (t) under the parameter vector Θ, corresponding to formula (28), represents the cumulative amount of real retweets of supporters; w is the sampling time, w = 0, 1, 2,... T.

[0130] To evaluate the numerical fitting effect, the Mean Absolute Percentage Error (MAPE) is introduced to evaluate the goodness of fit in this embodiment. The result of MAPE ranges from [0, +oo], and the result of 0% indicates a perfect model, and the result of MAPE greater than 100% indicates a poor model, therefore, the closer the result of this index to 0, the better the performance of the model, which can be expressed by formula (31)-(32):

[0131]

[0132] In the formula, m is the number of sample points in the support retweet data set of each case, n is the number of sample points in the anti-retweet data set of each case, y i(P / C) is the observed value of the i-th sample point, is the predicted value of y i(P / C) .

[0133] Figure 8 Figures (a)-(d) respectively show the numerical fitting results of the MF-SLPCI model of information propagation event case 1, case 2, case 3, and case 4, and the model parameter estimation results corresponding to the cases are shown in Table 3-4. Among them, the red star and the blue star respectively represent the real value of the support retweeter and the anti-retweeter, and the red solid line and the blue solid line respectively represent the simulation value of the cumulative retweet amount of the support retweeter and the cumulative retweet amount of the anti-retweeter calculated by the model. From the numerical simulation results, it can be seen that the curve of the case fitting by the MF-SLPCI model is almost coincided with the real data points. In addition, according to the calculation, the MAPE values of the four cases are shown in Table 4 and are close to 0. Therefore, it can be seen that the MF-SLPCI model constructed by the present application shows good fitting performance, and this result confirms the rationality of the three-dimensional network propagation mechanism under the influence of multiple factors formulated in the model, that is, the MF-SLPCI model can effectively simulate the information propagation process and has the ability to reproduce the dynamics of information propagation, which verifies the effectiveness and feasibility of the model.

[0134] Table 4: Fitting result evaluation index of cases 1-4 (unit: percentage)

[0135]

[0136] Firstly, from the fitting results of the four event development key times of Figure 8 , it can be concluded that the final size of the information propagation of the supporters in case 2 is the largest, and the final size of the information propagation of the supporters in case 4 is the smallest. The case is the development stage of the event, and the information content involves the education and gender-related information of the parties, which is easy to cause "one-sided" phenomenon on the social network platform, that is, the voice of supporting the school investigation is higher in this case, which will also cause a lot of attention and discussion of netizens. In addition, the final size of the information dissemination of the opponents in case 3 is greater than that of the other three cases, which is the climax stage of the event development. Correspondingly, the final size of the information dissemination of the opponents in case 3 is greater than that of the other three cases, and the final size of the information dissemination of the supporters of each information is greater than that of the opponents This is consistent with the view tendency analysis result of the real data in Figure 6 , which shows that the views of most netizens will be affected by the information published by the topic host, and the information dissemination will be maintained in line with the views of the information.

[0137] Table 5 Parameter estimation results of case 1

[0138]

[0139] Table 6 Parameter estimation results of case 2

[0140]

[0141] Table 7 Parameter estimation results of case 3

[0142]

[0143] Table 8 Parameter estimation results of case 4

[0144]

[0145] Secondly, from the environmental impact point of view, the number of environmental media reports n and the environmental media influence degree δ of the four case events are relatively close, that is, the media environment affecting the development of information dissemination is similar in the whole process, and there is no phenomenon such as reversal in the development process of the whole event. The overall information dissemination tone has been presented since the beginning of the event, but in case 4, the environmental media influence degree δ is smaller than that of case 1, case 2 and case 3, which is the end of the event development, and the influence of the media is also weakened accordingly; From the environmental influence degree factor τ, the size relationship is case 2> case 4> case 1> case 3, which is consistent with the reading volume and discussion volume of different stages of the topic.

[0146] From the individual attribute, the parameter fitting results of the four cases can find that the support view retention probability η P is greater than the opposition view retention probability η C , and the average forwarding probability λ of the supporters P are all greater than the average retransmission probability λ of the opponents C , which indicates that in the whole development process of the information dissemination event, netizens are more inclined to support the information released by the topic host and further disseminate the supporting opinions, which is consistent with the relevant discussion in the foregoing; from the immune rate θ of the lurkers, the size relationship is case 3 > case 1 > case 2 > case 4, which highlights the burst speed of each piece of information, that is, the burst speed of case 3 is the fastest. From the immune rates of the supporters and the retransmitters, the immune rate α P of the supporters in the four cases is greater than the α C of the opponents, and the values of α P and α C of case 4 are greater than those of the other cases, that is, the supporters are more likely to lose interest in the information and quit the information retransmission active state faster, and at the tail stage of the development of the information dissemination, the attractiveness to netizens is smaller, and it is easier to quit the information retransmission active state regardless of the opinion held.

[0147] In addition, from the network structure, the greater the parameter K m , the faster the burst speed of the information. From the fitting results of the four cases, it can be obtained that the size relationship of the values of K m is case 3 > case 1 > case 2 > case 4, which is also consistent with the corresponding relationship of the average transfer rate of the lurkers mentioned above. The value of K m in case 3 is the largest, and K m is the maximum value of the node degree in the network, representing the maximum number of users that a user can contact in the information dissemination process, which is an important parameter reflecting the characteristics of users in network information dissemination. The greater the parameter K m , the more fans or greater influence the information publisher in case 3 has compared with other information publishers, or other opinion leaders with greater influence retransmit the information after the information is released, so that the information can contact more users in the dissemination process, resulting in rapid burst of the information.

[0148] Based on the fitting results of the actual case events, the related parameters involved in the model and the physical factors they map are further analyzed. Specifically, the environmental influence degree factor τ, the environmental media influence degree δ, the environmental media number n, the support opinion retention probability η P , the average retransmission probability λ of the supporters λ P , the opposition opinion retention probability η C , the average retransmission probability λ of the opponents λ C , the initial value of the total quantity of susceptible population S0, the initial value of the total quantity of lurkers L0, the average transfer rate q of the lurkers, the total number of users N, and the maximum value of the node in-degree K m and the maximum out-degree L of the node m These thirteen parameters respectively affect the cumulative amount of information forwarding C. P(C) The influence of P(t) and instantaneous forwarding volume P(t) is as follows: Since this model is a general framework for studying different information dissemination cases, the parameter influence of the fitting results of different cases is similar when conducting the following analysis. This section only takes Case 3 as an example to conduct single parameter change analysis.

[0149] Environmental Impact Analysis

[0150] like Figure 9 As shown in (a)-(f), the parameters mapping the physical factors of environmental impacts—environmental impact degree factor τ, environmental media influence degree δ, and the number of environmental media n—are relative to C. P(C) and F P(C) The parameters have similar effects, and the same parameter has a greater impact on the cumulative and instantaneous forwarding volume for supporters than on the impact on opponents. Specifically, as the parameters τ, δ, and n increase, the instantaneous forwarding volume curve becomes steeper and the peak value is higher. Similarly, increasing these three parameters will accelerate the speed of information explosion and the cumulative forwarding volume will reach a larger final scale. Further comparative analysis of the impact of each parameter on the same indicator, analyzing Figures (a), (c), (e) and (b), (d), (f) respectively, shows that compared with parameter n, the number of media n has a greater impact on the information dissemination indicator. The final scale of information dissemination caused by a smaller change in its parameter value is significantly larger. Information dissemination peak P max / C max The greater the range of changes, the more important it is in terms of accelerating the speed of information explosion, expanding the scope of information coverage, and enhancing netizens' awareness of information. It is also an important factor that relevant departments cannot ignore in the process of information dissemination governance.

[0151] Overall, these factors primarily affect the peak volume, scale, and speed of information dissemination. Therefore, collaboration with mainstream media and new media platforms is crucial during the development of relevant information dissemination events. Through news reports and commentaries, public opinion can be guided, positive publicity promoted, and negative impacts reduced. Furthermore, relevant publicity and educational activities can enhance public media literacy, enabling them to view and disseminate information more rationally and reducing the spread of misinformation.

[0152] Analysis of the influence of individual characteristics

[0153] like Figure 10 As shown in (a)-(f), the parameters that map the physical factors related to the individual characteristics of netizens and support the probability of retaining the viewpoint η P The average forwarding probability λ of supporters P , the probability of holding the opposing view η C , the average forwarding probability of the opposing group λ C , the initial value of the total quantity of susceptible group S0, the initial value of the total quantity of latent group L0, the average transfer rate of the latent group q, the total quantity of users N P(C) and F P(C) have similar influences. Specifically, as the parameters η P(C) , λ P(C) , S0, and L0 increase, the instantaneous forwarding quantity curve of the supporters (opponents) becomes steeper, the information propagation peak P(C) max becomes higher, and the increase of the two types of parameters will accelerate the speed of the information explosion and make the cumulative forwarding quantity of the supporters (opponents) reach a larger final scale. In addition, the increase of the parameter S0 will also advance the time t max at which the instantaneous forwarding quantity of the users reaches the peak. Compared with the above parameters, the parameter N has a relatively large difference in terms of the degree and effect of influence, that is, the parameter N has a relatively small negative influence on C P(C) and F P(C) , and as N increases, C P(C) and F P(C) will decrease, and the increase of the size of the crowd will also increase the difficulty of information diffusion. From the parameter q, the average transfer rate of the latent group, it mainly influences that as the parameter increases, the time at which the instantaneous forwarding quantity decreases to 0 is earlier, that is, the time at which the final scale of information propagation reaches the stable state is earlier. In the comparative analysis of the same parameters in the same range, compared with the opposing group, the influence of the parameters on the change of the cumulative forwarding quantity and the instantaneous forwarding quantity of the supporters is greater, and the speed of the information explosion in the same time is also relatively greater, which is consistent with the above conclusion. Further comparative analysis of each parameter is made on (a), (c), (e), (g), (i), (k) and (b), (d), (f), (h), (j), (l) in Figure 10 . For the influence on the same index, the influence of the parameters η P(C) and λ P(C) on the cumulative forwarding quantity and the instantaneous forwarding quantity is greater, followed by S0, L0, and q, and N is the least, which also provides a certain direction for information propagation management.

[0154] Overall, these factors primarily influence the peak and scale of information dissemination, the speed of information explosion, the time it takes for the instantaneous forwarding volume to reach its peak, and the time it takes for the final scale of information dissemination to stabilize. Therefore, information dissemination governance can be approached from these factors. Based on the susceptibility of different groups and their probabilities of retaining opinions and forwarding, targeted propaganda strategies can be developed, and information can be delivered in a targeted manner to increase the coverage and influence of positive information. Furthermore, based on the forwarding probabilities of different groups and the number of susceptible individuals, information dissemination can be managed in a tiered manner, allowing for the timely detection and early warning of high-risk information dissemination events, and the implementation of corresponding countermeasures. Through online and offline activities, public media literacy can be improved, enabling them to make rational judgments and share information, reducing the number of susceptible individuals, and thus allowing for more scientific and effective management and guidance of information dissemination.

[0155] Network Structure Impact Analysis

[0156] like Figure 11 As shown in (a)-(d), the maximum in-degree K of the node m and the maximum out-degree L of the node m For F P(C) and C P(C) They have similar effects. With K m and L m As the degree increases, the instantaneous forwarding curve becomes steeper, the peak value becomes higher, and the burst speed increases. Similarly, the final scale of accumulated forwarding users also increases with the increase of these two parameters. That is, the larger the out-degree, the more information can be spread to other nodes, and it usually has a greater propagation capacity in the network, and can more effectively spread information to a wider group. The higher the in-degree, the stronger the ability to receive information from other nodes, and it can usually obtain new information in the network faster, thus affecting whether it can become a source or a disseminator of information. Further, by analyzing Figures (a) and (c), and Figures (b) and (d) respectively, the parameter K can be derived. m With L m In comparison, K m The changes in parameters have a greater impact on the cumulative forwarding volume and the instantaneous forwarding volume, and the final scale of the cumulative volume and the peak value of the instantaneous forwarding volume have a larger range of variation. These patterns are of great significance when analyzing or designing information dissemination strategies.

[0157] Overall, these factors mainly affect the peak, scale, and speed of information dissemination. Relevant departments can start with these factors, analyze in-degree and out-degree, monitor the information dissemination of key nodes in real time, strengthen supervision of nodes with high in-degree and high out-degree, require them to exercise self-discipline, avoid spreading false information, malicious rumors, and other harmful information, promptly detect and warn of events that may cause information dissemination risks, and take corresponding measures to intervene and deal with them.

[0158] As can be seen from the above examples, the information propagation analysis method based on the three-dimensional network information propagation dynamics model provided by the application constructs a general framework of a three-dimensional network information propagation dynamics model based on multiple elements driven by internal and external driving forces, divides various factors affecting information propagation into layers, divides them into social network layer, environmental time sequence layer, psychological layer and viewpoint layer, quantifies the internal and external driving forces distributed in each layer into nested coupling functions and embeds them into the differential equation representing the information propagation dynamics of the social network layer, and takes the time sequence of the environment layer as the main line, performs numerical fitting on the cumulative amount of forwarding information of case information users at multiple key time points of the development process of a certain information propagation event, verifies the effectiveness and rationality of the model, and shows that the constructed MF-SLPCI model can capture the dynamic process of information propagation and evolution. Further, an information propagation index is constructed to measure the propagation and evolution effect, and then a single parameter variation analysis experiment is performed to analyze the information propagation influencing factors from the aspects of environmental influence, individual characteristic influence and network structure influence, fully showing that the internal and external driving force factors are closely related to the dynamic propagation and evolution process of information.

[0159] In summary, the application fuses internal and external driving forces and quantifies them as time-varying parameters coupled and nested, constructs an information propagation dynamics model based on multiple elements, and analyzes the physical factors mapped by the model parameters based on the propagation characteristics of network information propagation evolution, reveals the objective action law of multiple elements on information propagation and information propagation evolution, so that relevant departments can develop corresponding measures to more targetedly govern social network information propagation and reduce the propagation of bad information.

[0160] Corresponding to the above information propagation analysis method based on the three-dimensional network information propagation dynamics model, the application also provides an information propagation analysis system based on the three-dimensional network information propagation dynamics model, which is used for information propagation analysis by using the information propagation analysis method based on the three-dimensional network information propagation dynamics model as described above, and includes:

[0161] An original propagation data acquisition unit is configured to determine key time points in the development process of a selected information propagation event, and extract hot topic event information of the information propagation event at the key time points and forwarding information in a forwarding area of the hot topic event information as original propagation data, wherein the forwarding information includes forwarding text and forwarding time of forwarding users under the hot topic event information.

[0162] a data preprocessing unit configured to preprocess the original propagation data to obtain propagation effective data in the original propagation data;

[0163] a simulation analysis unit configured to simulate a group state transition and an information propagation event development evolution dynamic process under the three-dimensional network of the social media platform based on the propagation effective data and a preset three-dimensional network information propagation dynamics model.

[0164] For a more specific embodiment of the information propagation analysis system based on the three-dimensional network information propagation dynamics model of the present application, reference can be made to the foregoing description of the information propagation analysis method based on the three-dimensional network information propagation dynamics model, which will not be described one by one in detail here.

[0165] The information propagation analysis method and system based on the three-dimensional network information propagation dynamics model of the present application are described above with reference to the accompanying drawings by way of example. However, those skilled in the art should understand that various improvements can be made to the information propagation analysis method and system based on the three-dimensional network information propagation dynamics model of the present application described above without departing from the content of the present application. Therefore, the protection scope of the present application should be determined by the content of the appended claims.< / l> < / l>

Claims

1. An information propagation analysis method based on a stereoscopic network information propagation dynamics model, for simulating the state transition of netizen individuals and the evolution trend of information propagation in a network information propagation event based on a preset stereoscopic network information propagation dynamics model, wherein the structure of the stereoscopic network comprises: a social network layer for representing the directed complex network structure of netizen nodes participating in the information propagation process through a set of a limited number of information propagation subject nodes and directed edge relationships between the nodes, wherein the netizen individuals on each social platform are nodes, and the edges between the nodes represent the propagation of information between users participating in the stereoscopic network information propagation, and the edges pointing to the user nodes represent the influence of other users on the users, and the edges diverging from the user nodes represent the ability of the users to propagate information outward; an environment layer for representing the information propagation environment under the beginning, development, climax and end of an information propagation event through a set of a limited number of environment type nodes and directed edge relationships between the nodes; wherein the time sequence relationship of the environment information is represented by the directed edges of the environment layer, and the directed connection between the environment nodes represents the continuous change of the environment; a psychological layer for representing the psychology of the netizens to the propagation of the information propagation event and the transformation relationship between the psychological types through a set of a limited number of psychological type nodes and directed edge relationships between the nodes, wherein the psychological state transition of the psychological layer is influenced by the environment layer; a viewpoint layer for representing the transformation of the viewpoints of the netizens in different periods of the information propagation event through a set of a limited number of viewpoint type nodes and directed edges between the nodes, wherein the viewpoints of the viewpoint layer are divided into positive and negative viewpoints according to the viewpoint theme and emotional tendency. The method comprises: determining the key time points in the development process of the selected information propagation event; extracting the hot topic event information of the information propagation event at the key time points and the forwarding information in the forwarding area of the hot topic event information as original propagation data; wherein the forwarding information comprises the forwarding text and forwarding time of the forwarding users under the hot topic event information; preprocessing the original propagation data to obtain the propagation effective data in the original propagation data; simulating the group state transition and information propagation event development evolution dynamic process under the stereoscopic network of the social media platform based on the propagation effective data and the preset stereoscopic network information propagation dynamics model. The super-edge of the three-dimensional network The stereoscopic network information propagation dynamics model is: The subject Under the joint action of the external driving force of the information dissemination environment And the internal driving force of the psychological state Published views Finally 、 、 Affecting the information dissemination subject The change of the dissemination state in the social network layer; the three-dimensional network information dissemination dynamics model takes the time sequence relationship of the event development process of the environment layer as the main line, takes the overall other factors of the environment layer as the external driving force, takes the nested influence of the psychological layer and the view layer as the internal driving force, and is based on the general information dissemination scene formed by the dissemination state evolution process of the user node in the social network layer under the joint action of the internal and external driving forces In the stereoscopic network, the average in-degree is equal to the average out-degree. In the stereoscopic network information propagation dynamics model, the social network layer, the environment layer, the psychological layer and the viewpoint layer all influence the propagation state of the netizens in the process of state transition of the netizens, and the parameters of the stereoscopic network are set as shown in the following formula: In the stereoscopic network, during the incubation period, an information propagation analysis method based on the stereoscopic network information propagation dynamics model of any one of claims 1-9, comprising: ​ 2. The information propagation analysis method based on the stereoscopic network information propagation dynamics model according to claim 1, wherein, Assuming that information is spread in a closed stable environment, only considering the population that can be reached in the process of information spread itself, the node state in the three-dimensional network information propagation dynamics model includes susceptible state S, latent state L, support state P, opposition state C and immune state I; the in-degree of the node of the three-dimensional network is defined and the out-degree of the node According to the in-degree and the out-degree of different nodes in the three-dimensional network, the population in each state is classified, which includes: representative at the moment, the total number of individuals in a susceptible state with in-degree , out-degree ​ represent at the moment, the total number of individuals in the latent state with in-degree , out-degree ; represent at the moment, the total number of individuals in the support forwarding state is , the in-degree is , and the out-degree is represent at the moment, the total amount of individuals in the anti-forwarding state with in-degree , out-degree ; represent at the moment, the total number of individuals in the immune state with in-degree , out-degree ; The sum of the total number of moments for each population is a function of time The sum of the total number of moments for each population is a function of time ; ​ （2） （3） （4） （5） （6） （7） （8） （9） wherein, represents the state of an individual from state X migrates into state Y a stereoscopic network influence parameter, represents the average migration rate, represents the average migration rate, represents the average transfer rate of a latent, represents the in-degree of a node in the stereoscopic network, represents the out-degree of a node in the stereoscopic network, represents the average immunity rate of a pro-forwarder, represents the average immunity rate of an anti-forwarder; The initial values of each state are defined as: = 、 、 , , ; denotes the conditional probability that any edge in the directed network is emitted by a node with "degree" of and directed to a node with "degree" of denotes the probability that a node with "degree" of is connected to a user in the support forwarding state at time denotes the probability that a node with "degree" of is connected to a user in the oppose forwarding state at time denotes the probability that a node is connected to a user in the forwarding state.​​​ 3. The information propagation analysis method based on the three-dimensional network information propagation dynamics model according to claim 2, wherein, Assuming that the stereonet is an uncorrelated stereonet, the conditional probability is only related to the degree of the node with upstream indegree outdegree of ​ （10） wherein, and denote the in-degree and out-degree of different nodes, denote user nodes with different in-degree and out-degree , denote the joint probability distribution of nodes in the network, equal to the ratio of the total number of nodes in the network with in-degree and out-degree to the total number of nodes in the network, denote the average degree of the network, the degree distribution of any point in the network being: ​​​ （11）。 4. The information propagation analysis method based on the three-dimensional network information propagation dynamics model according to claim 3, wherein, ​ 5. The information propagation analysis method based on the three-dimensional network information propagation dynamics model according to claim 4, wherein, ​ （15） （16） （17） （18） （19） （20） （21） （22） （23） （24） （25）； wherein, represents an environmental impact factor, represents an environmental media impact, represents an environmental media number, represents a supporting opinion retention probability, represents a supporting average retweet probability, represents an opposing opinion retention probability, represents an opposing average retweet probability.

6. The information propagation analysis method based on the three-dimensional network information propagation dynamics model as claimed in claim 5, wherein, In the network, a degree of a user in a vulnerable state per unit of time contacts a user in a supportive forwarding state is When users contact a supporter / opponent, where: The susceptible person will maintain a constant focus on the information from state to state; susceptible will support and forward information consistent with the original information, from state to state; a susceptible person will forward the information with an opposing attitude that is inconsistent with the original information, from state to state; The rest of the susceptible persons are not interested in the information, have one susceptible person moves from state to state.

7. The information propagation analysis method based on the three-dimensional network information propagation dynamics model according to claim 6, wherein, ​ There is an increase in the willingness of lurkers to propagate information resulting in a retweeting behavior, a lurker moves from a lurker state to an active state; There are one incubator from state to state; the remaining The lurking agent lost interest in the information, that is, from State transition to state.

8. The information propagation analysis method based on the three-dimensional network information propagation dynamics model according to claim 7, wherein, In the aforementioned three-dimensional network, for a forwarder, when supporter A / opponent B is influenced by fellow supporter A / opponent B, they will respectively have A supporter from State transition to state One opponent from State transition to In this situation, another group of supporters and opponents will respectively... , The quantity from State transition to Status, the rest / Supporters / opponents, having passed their peak of exposure, are no longer able to influence others. State transition to state.

9. The information propagation analysis method based on the stereoscopic network information propagation dynamics model according to claim 8, wherein, In the propagation dynamics model of the stereoscopic network information, the forwarding cumulative amount is directly obtained from the network propagation platform, and the degree is The change rate of the forwarding cumulative amount of the supporting forwarding user and the opposing forwarding user with time is: （26） （27） The cumulative amount of retweets by supporters and opponents and are respectively: （28） （29）。 10. An information propagation analysis system based on a three-dimensional network information propagation dynamics model, characterized by, ​ The original propagation data acquisition unit is configured to determine a key time point in the development of the selected information propagation event, and extract hot topic event information of the information propagation event at the key time point and forwarding information in a forwarding area of the hot topic event information as original propagation data, wherein the forwarding information includes forwarding text and forwarding time of a forwarding user under the hot topic event information; The data preprocessing unit is configured to preprocess the original propagation data to obtain propagation effective data in the original propagation data. The simulation analysis unit is configured to simulate group state transition and information propagation event development evolution dynamic processes under the three-dimensional network of the social media platform based on the propagation effective data and a preset three-dimensional network information propagation dynamics model.

Images