Online viewpoint evolution simulation method and device based on offline event incentive model

Abstract

This disclosure provides a method, apparatus, computing device, and storage medium for simulating online opinion evolution based on an offline event incentive model. The method includes: acquiring social media data related to public opinion events, extracting the user interaction network structure and user opinion values ​​at different time points during the development of public opinion; modeling the sequence of real events as a Hawkes point process based on the public opinion event data to obtain the Hawkes function of the real events; constructing an opinion dynamics model based on offline event incentives according to a model representing the user opinion interaction mechanism, social norm regulation factors, and the Hawkes function, and tuning parameters using the Monte Carlo method; inputting the initial user opinions obtained from the real data into the opinion dynamics model, and iteratively calculating until the opinion values ​​of all users in the network converge to a stable state. This invention can accurately model the impact mechanism of offline events on the development of online public opinion to a great extent, realizing the simulation of user opinion evolution on social media platforms.

Description

Technical Field

[0001] This invention relates to the field of social network application technology, and specifically to a method, apparatus, computing device, and storage medium for simulating the evolution of social media opinions based on an offline event incentive model. Background Technology

[0002] With the rapid development of the internet, people are increasingly interacting and communicating on social media platforms. Real-world events can be easily published on social media, sparking widespread discussion and influencing the evolution of public opinion.

[0003] Current research on the simulation of opinion evolution mainly focuses on opinion dynamics, which aims to understand and explain the mechanisms by which opinions are formed, evolved, and spread among people. Existing research on opinion dynamics on social media primarily considers specific nodes in the network and the unique interaction mechanisms of social media to simulate and explain phenomena such as echo chambers and spirals of silence. However, these methods neglect the incentive and altering effects of real-world events on the evolution of online opinions, as well as the restrictive influence of platform regulations on opinion extremism, leading to insufficient accuracy in the simulation results. Summary of the Invention

[0004] In view of this, the present invention is proposed to provide an online viewpoint evolution simulation method based on an offline event-driven model that overcomes or at least partially solves the above-mentioned problems, comprising:

[0005] Step S101: Obtain social media data about the public opinion event and construct a user interaction network structure; quantify the opinion tendencies in users' historical posts to obtain the opinion values ​​of users at different time periods during the development of public opinion;

[0006] Step S102: Based on the public opinion event data, model the sequence of real events as a Hawkes point process to obtain the Hawkes function of the real events;

[0007] Step S103: Based on the model representing the user opinion interaction mechanism, the social norm regulation factor that guides opinions to align with public opinion, and the Hawkes function of the real event, construct an opinion dynamics model based on offline event incentives; determine the parameters of the opinion dynamics model using the Monte Carlo method based on the user interaction network structure and user opinion values ​​in the public opinion event.

[0008] Step S104: Input the initial user opinion values ​​obtained from the real data into the opinion dynamics model with determined parameters, and iterate until the opinion values ​​of all users in the user interaction network structure converge to a stable state, thus completing the online opinion evolution simulation.

[0009] Preferably, the quantification of user opinion tendencies in historical posts yields user opinion values ​​at different time points during the development of public opinion:

[0010] Users' social media posts are encoded into vectors, and then normalized to obtain opinion values ​​by measuring the distance between them and pre-selected vectors of extreme opinions.

[0011] Preferably, the vector distance between two texts is obtained by using the average cosine similarity between the word vectors in the text.

[0012] Preferably, step 102 involves: modeling the real time series as a self-excitation point process, with the excitation function employing an exponential decay function; then the Hawkes function of the real event is expressed as:

[0013]

[0014] Where, λ base >0 represents the base strength, and t represents the current time. p Let t represent different times in the development process before t, where constants a, b > 0.

[0015] Preferably, in step 103, during the process of determining the parameters of the opinion dynamics model using the Monte Carlo method, considering that user opinion interaction follows a confidence interval mechanism, the confidence interval is further used as a parameter tuning object to adjust the interaction weights h between interaction nodes in the user interaction network structure to be substituted into the opinion dynamics model. ij Perform a second zeroing check: If the difference in opinions between two interacting nodes is not within the confidence interval, no interaction will occur, and the interaction weight h between the two interacting nodes will be set. ij =0, and then substitute it into the viewpoint dynamics model.

[0016] Preferably, step 103 is:

[0017] First, based on the FJ model representing the interaction mechanism of user opinions and the social norm regulation factor that guides opinions towards the general public, a model of opinion evolution under the constraints of social norms is constructed as follows:

[0018]

[0019] Where, x i (t) represents the opinion value of user i, which changes with time t; K i Ω represents the social norm regulation coefficient for user i; α represents the neutral opinion value; α is the regulation parameter, the larger the α, the better the regulation effect; μ i h is the influence coefficient of user i's opinion. ij This represents the weight of the edge connecting user i and user j in the user interaction network structure;

[0020] The expression for the viewpoint dynamics model is:

[0021] dx(t)=[(I n -Λ-K-ω)x(0)+Ku+Λh(x(t))]dt+ωdλ * (t)

[0022] Among them, I n Let denot be the identity matrix, and Λ denote the diagonal matrix composed of sensitivity coefficients μ; Let y represent the feedback control vector of a social media platform for each user's opinion, and let y be the feedback control factor for user k. k (t)=-tanh(αx k (t)); K=diag(K1, K2,...,K n ) represents the social norm regulation coefficient matrix; ω = diag(ω1, ω2, ..., ω n ) represents the matrix showing the degree of influence of events occurring in the real space on different users, for any coefficient ω i ∈[0,1];λ * (t) represents the conditional strength function of an offline event in real space, i.e., the Hawkes function expression of a real event; dλ * (t) represents the conditional intensity variable within the time window [t, t+dt); h(x) represents h ij The edge weight matrix is ​​formed.

[0023] Preferably, the neutral opinion value Ω is taken as the average of the maximum and minimum values ​​of each user opinion in the user interaction network structure.

[0024] The online viewpoint evolution simulation device based on an offline event-driven model provided by this invention includes:

[0025] The public opinion event data processing and extraction module is used to acquire social media data about public opinion events, construct a user interaction network structure, quantify the opinion tendencies in users' historical posts, and obtain the opinion values ​​of users at different time periods in the development of public opinion.

[0026] The offline event activation function module is used to model the sequence of real events as Hawke's point processes based on public opinion event data, and obtain the Hawke's function of the real events.

[0027] The dynamics model parameter tuning module is used to construct a viewpoint dynamics model based on offline event incentives, according to the model representing the user viewpoint interaction mechanism, the social norm regulation factor that regulates viewpoints to align with public opinion, and the Hawkes function of the real event; and to determine the parameters of the viewpoint dynamics model using the Monte Carlo method based on the user interaction network structure and user viewpoint values ​​in the public opinion event.

[0028] The opinion evolution simulation module is used to input the initial user opinion values ​​obtained from real data into the opinion dynamics model with defined parameters, and iteratively calculate until the opinion values ​​of all users in the user interaction network structure converge to a stable state, thus completing the online opinion evolution simulation.

[0029] According to another aspect of the present invention, a computing device is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0030] The memory is used to store at least one executable instruction, which enables the processor to perform the operation corresponding to the above-described online viewpoint evolution simulation method based on an offline event-driven model.

[0031] According to another aspect of the present invention, a computer storage medium is provided, wherein at least one executable instruction is stored therein, the executable instruction causing a processor to perform the operation corresponding to the above-described viewpoint simulation evolution method based on an offline event-driven model.

[0032] Beneficial effects:

[0033] (1) This invention, taking into account the influence of offline events on social media platforms, uses Hawkes processes to describe the sequence of offline events and establishes an opinion evolution model that conforms to the characteristics of social media platforms. This model can predict the trend of user opinions in social media public opinion events, facilitating the control of public opinion dissemination on social media and contributing to the construction of a healthier and more harmonious online community space. Furthermore, this invention also captures the characteristic that extreme content on social media is restricted by platform norms, adding social norm regulation factors to the opinion evolution model, thereby capturing the characteristics of opinion evolution in user interactions on social media platforms from both online and offline perspectives. Compared with existing methods, this invention increases the dimension of influencing factors in the opinion evolution process, retains the core mechanism of opinion interaction, and achieves accurate fitting of the development process of social media public opinion events.

[0034] (2) Online social media sentiment changes as offline events occur and progress. To represent this impact, the influence of real events on social media platforms is simulated to gradually decrease. In a preferred embodiment, the exponential decay function of the Hawkes process is adopted. The exponential decay function can describe the increase in intensity as offline events occur, while decreasing over time, which is consistent with the changing trend of public opinion discussion on social media platforms.

[0035] (3) In a preferred embodiment, a hyperbolic tangent function is used to model the social norm regulation factor. The nonlinear shape of the function ensures that individuals will adjust in the direction of regulation and gradually approach a neutral viewpoint, which can describe the influence of social norms on extreme viewpoints.

[0036] (4) In a preferred embodiment, consideration is further added to the user opinion interaction following a confidence threshold mechanism, and the user interaction weight h is adjusted accordingly. ij A second judgment is made: if the difference in opinions between the two interacting nodes is not within the confidence interval, then no interaction will occur. Therefore, h... ij The value should be 0, thus describing how users with vastly different viewpoints find it difficult to communicate and influence each other, reproducing the discussion atmosphere in the social media public opinion environment, and more accurately fitting the development process of social media public opinion events. Attached Figure Description

[0037] Figure 1 This invention provides a flowchart of a viewpoint simulation evolution method based on an offline event-driven model, according to an embodiment of the present invention.

[0038] Figure 2 This diagram illustrates the text opinion value extraction process provided in an embodiment of the present invention.

[0039] Figure 3 This diagram illustrates the structure of a viewpoint simulation evolution device based on an offline event-driven model, according to an embodiment of the present invention.

[0040] Figure 4 A schematic diagram of the structure of a computing device provided in an embodiment of the present invention is shown. Detailed Implementation

[0041] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0042] Example 1

[0043] Figure 1 The flowchart of a social media opinion evolution simulation method based on an offline event incentive model according to the present invention is shown, as follows: Figure 1 As shown, the method includes the following steps:

[0044] Step S101: Obtain relevant social media data of the public opinion event and extract the user interaction network structure from it; quantify the opinion tendencies in users' historical posts and obtain the opinion values ​​of users at different time periods in the development of public opinion.

[0045] This embodiment employs an opinion dynamics model based on offline event incentives, simulating the evolution of user opinion values ​​using a real-world interactive network structure. Specifically, the real-world interactive network structure refers to a network structure built based on relationships such as comments, reposts, and follows. Each node represents a user, and each edge represents an interaction relationship. Edges are all-inclusive, and their weights are determined based on interaction frequency and interaction type (e.g., reposts, comments).

[0046] The approach to quantifying user opinion values ​​is as follows: User social media posts are encoded into vectors, and the opinion value is obtained by normalizing the vectors based on their distance from pre-selected vectors of extreme opinions. The vector distance between two texts can be calculated using the mean cosine similarity between the word vectors in the text.

[0047] like Figure 2 The diagram illustrates the specific process of quantifying user opinion values. As shown, internet information text data is often unstructured and contains special characters such as punctuation, symbols, and emoticons, thus requiring preprocessing. First, noisy words are removed using a stop word list to improve the accuracy of subsequent processing. Second, the text is segmented based on a corpus to identify key words that constitute the semantics of sentences. Then, these key words are encoded, and word2vec (a word-to-vector mapping model) is used to learn the distributed representation of words in a large corpus, obtaining semantic and syntactic information to obtain word vectors. The text containing pre-selected, manually chosen extreme opinions is processed in the same way to obtain the embedding vectors of all segmented words, and the average embedding vector of the extreme opinion content is obtained by averaging. Other content in related topics on social media platforms is processed in the same way to obtain average vectors, and then the cosine similarity is calculated to quantify the vector distance between it and the average embedding vector of the extreme opinion content. For two vectors u and v, the cosine similarity is calculated as follows:

[0048]

[0049] Where u·v represents the inner product of u and v, and ||u|| and ||v|| represent the moduli of u and v, respectively.

[0050] The vector distance between texts can be obtained by calculating the mean cosine similarity between each word vector. Assume text A and text B are represented by a1, a2, ..., a1, ..., a2, ..., a3, ..., a4, ... N and b1, b2, ..., b N′ Composition, where N and N′ are the number of words in the two texts respectively, then the average cosine similarity between text A and text B is:

[0051]

[0052] Finally, by normalizing these distances, the opinion value for each piece of content is obtained. The value range of opinion value is set to -1 to 1, where more positive sentiment is closer to 1, neutral sentiment is closer to 0, and negative sentiment is closer to -1.

[0053] Step S102: Based on the public opinion event data, model the sequence of real events as a Hawkes point process, determine the values ​​of relevant parameters in the modeling, and obtain the Hawkes function of the real events.

[0054] Online social media sentiment changes as offline events occur and unfold. To represent this impact, this invention describes offline events as event sequences {t1, t2, ...}, modeled using Hawkes point processes.

[0055] Specifically, the number of events that occurred before time t is expressed as:

[0056]

[0057] Among them, t k This indicates the time when event p occurred. This is an indicator function, indicating that event i occurred before time t. Historical events. The counting process N(t) satisfies:

[0058]

[0059] in, Let m be the number of new events added between two consecutive moments, and o(h) represent the minimum probability term; λ * (t) is the conditional intensity function, in the form:

[0060]

[0061] Where, λ base >0 represents the base strength, and φ: (0, ∞) → [0, ∞) is the excitation function, representing the influence of the triggering event on the occurrence of subsequent events.

[0062] For an observed sequence of events {t1, t2, ...}, the conditional strength function of the Hawkes process is:

[0063]

[0064] To model the gradual decay of the influence of real-world events on social media platforms, the activation function φ(·) adopts an exponential decay function, with φ(t) = ae -bt Where constants a and b > 0, then we have:

[0065]

[0066] The above equation shows that the system strength increases by an amount of 'a' each time, and then decays at a rate of 'b' with each event. The real time series is modeled as a self-excited process, where each event increases the probability of another similar event occurring.

[0067] From a series of related real events, key time points that significantly influence online public opinion are manually extracted and represented as time series point processes. Based on likelihood estimation, the Hawkes form of the real events is obtained.

[0068] Specifically, the likelihood function of the Hawkes process is in the form of:

[0069]

[0070] Since the condition function of the Hawkes process used in this method is an exponentially decaying function, for [0, t]... p Events on ], let The likelihood function of the Hawkes process in this interval can be written as:

[0071]

[0072] [0, t] p Discretization, that is, dividing it into [0, t1], [t1, t2], ..., [t...]. p-1 , t p Therefore, we have:

[0073]

[0074] The likelihood function of an exponential Hawkes process is:

[0075]

[0076] The parameters in the Hawkes function can be calculated using the above formula based on actual data.

[0077] Step S103: Based on the model representing the user opinion interaction mechanism, the social norm regulation factor that guides opinions to align with public opinion, and the Hawkes function of the real event, construct an opinion dynamics model based on offline event incentives; and determine the parameters of the opinion dynamics model using the Monte Carlo method based on the user interaction network structure and user opinion values ​​in the public opinion event.

[0078] This embodiment constructs an opinion dynamics model based on three considerations: user opinion interaction mechanisms, social norm regulation factors, and the occurrence and progress of offline events. Furthermore, during model parameter tuning, consideration was added regarding the user opinion interaction conforming to a confidence threshold mechanism, resulting in a comprehensive opinion dynamics model based on offline event incentives.

[0079] (1) User opinion interaction mechanism

[0080] The user opinion interaction mechanism is based on the Friedkin-Johnsen (FJ) model, where users adjust their opinions based on the views of others after interaction. The differential equation of the FJ model is as follows:

[0081]

[0082] Where, x i μ represents the opinion value of individual i, which changes with time t. i h is the influence coefficient of individual i's opinion. ij This represents the weight of the edge connecting user i and user j in the user interaction network structure. ∈ indicates a random perturbation, which generally follows a Poisson distribution.

[0083] (2) Social norm regulation factors

[0084] The evolution of opinions on social media platforms is influenced not only by offline events but also by platform norms. The existence of echo chambers and other phenomena makes social media platforms prone to opinion polarization, but extreme expressions of opinion are restricted by the platform. Platform policies tend to encourage more rational and neutral discourse; therefore, the evolutionary mechanism of opinion dynamics includes the regulation of extreme viewpoints, which can be termed a social norm regulatory factor.

[0085] The existence of social norm regulation factors reduces group polarization. The regulation factors and viewpoints deviate in opposite directions, and under the constraint of these factors, the evolution of viewpoints still satisfies the viewpoint value range. Let the regulatory intensity of social norms be y(t), and the regulatory amplitude be K∈(0,1). The regulatory mechanism is described using the hyperbolic tangent function tanh(·), where α>0 is the control coefficient of the sigmoid regulatory function. Therefore, the equation for the regulation factor is:

[0086]

[0087] This non-linear shape ensures that individuals will adjust according to the controlled direction of viewpoint evolution. This invention sets the viewpoint evolution direction to gradually approach the neutral viewpoint Ω. Therefore, based on the user viewpoint interaction mechanism in (1) and the social norm constraints in (2), the viewpoint evolution model under social norm constraints is as follows:

[0088]

[0089] Among them, K i Ω represents the social norm control coefficient for user i; Ω represents the neutral opinion value, which is the average of the maximum and minimum opinion values ​​for x. i ∈[-1, 1], Ω=0; the larger the control parameter α, the better the control effect.

[0090] (3) The occurrence and progress of offline events

[0091] The occurrence and progress of offline events are expressed using Hawkes functions.

[0092] Combining (1), (2), and (3), the differential equation form of the viewpoint dynamics model based on offline event incentives is obtained as follows:

[0093] dx(t)=[(I n -Λ-K-ω)x(0)+Ku+Λh(x(t))]dt+ωdλ * (t);

[0094] Among them, I n Let denot be the identity matrix, and Λ denote the diagonal matrix composed of sensitivity coefficients μ; Let y represent the feedback control vector of a social media platform for each user's opinion, and let y be the feedback control factor for user k. k (t)=-tanh(αx k (t)); K=diag(K1, K2,...,K n ) represents the social norm regulation coefficient matrix; ω = diag(ω1, ω2, ..., ω n ) represents the matrix showing the degree of influence of events occurring in the real space on different users, for any coefficient ω i ∈[0,1];λ * (t) represents the conditional strength function for the occurrence of offline events in real space, dλ * (t) represents the conditional strength variable within the time window [t, t+dt); h(x) represents the edge weight matrix.

[0095] The opinion dynamics model based on offline event incentives utilizes social norm regulation factors to reflect the influence of the average public opinion on users, causing user opinions to converge with the public opinion. On the other hand, it incorporates the influence of offline events in the real space on the evolution of opinions.

[0096] After determining the model form, the Monte Carlo method based on minimum average error is used for parameter tuning, according to the network structure and user opinion values ​​in the public opinion event. In the above formula, Λ, K, ω, and α are all parameters to be tuned. During parameter tuning, the user interaction network structure in the public opinion event (where the weights h) is considered.ij ) and user opinion value x i Substitute the parameters into the model and use the Monte Carlo method to determine the parameters of the viewpoint dynamics model.

[0097] During parameter tuning, the determination of the h(x) value was further considered. Considering that user opinion interaction follows a confidence interval mechanism, if the difference in opinions between two interacting nodes is not within the confidence interval, no interaction will occur. Therefore, h... ij It should be 0. Then, further settings will use the confidence interval as the parameter tuning target, adjusting the interaction weights h to be substituted into the viewpoint dynamics model. ij Perform a second zeroing check: If the difference in opinions between two interacting nodes is not within the confidence interval, no interaction will occur, and the interaction weight h between the two interacting nodes will be set. ij =0, and then substitute it into the opinion dynamics model. Through the parameter tuning process, the upper and lower limits of the confidence interval for user opinion interaction are determined.

[0098] The parameter tuning process includes two phases: training and testing. Early user opinion values ​​from the real data are used as the training set to tune the model. Later opinion values ​​are used as the test set to check the model's accuracy. Specifically, first, a set of parameter combinations is randomly generated in the parameter space according to the parameter value range defined in the model. For each set of parameters, the mean absolute error (MAE) of the model is calculated using the training set data, and the calculation is repeated to verify the results. The mean absolute error is a commonly used performance evaluation metric for regression models, used to measure the average difference between the model's predicted values ​​and the true values. Let y be the true values ​​of n samples in the training set. i The model's prediction values ​​for these samples are The mean absolute error is defined as:

[0099]

[0100] A smaller MAE indicates a better fit of the model to the samples. This method iterates through multiple iterations to gradually optimize the parameters until the number of iterations reaches its maximum value, or until a certain parameter combination has a small MAE on the validation set. The parameter combination with the smallest MAE is then selected as the model's computational parameters.

[0101] Step S104: Input the user's initial state (opinion value at the first moment) obtained from the real data into the opinion dynamics model based on offline event incentives, and iterate until the opinion values ​​of all users in the network converge to a stable state.

[0102] Having obtained the interactive network structure of the public opinion event and users' opinion values ​​at different times from step S101, the edge weight matrix in the model is determined based on the interactive network structure, and the users' opinion values ​​at the initial time are taken as the initial values ​​of the model. If there is a user whose opinion value data is missing at the initial time, interpolation is used to fill in the missing opinion value data so that the data for all users at the initial time can be obtained.

[0103] The Hawkes activation function of the real event is obtained from step S102, and the optimal parameters of the viewpoint dynamics model based on offline event incentives are obtained from step S103. The edge weight matrix and initial viewpoint values ​​determined in S101 are input into the viewpoint dynamics model for iterative calculation until the viewpoint values ​​of all users in the network converge to a stable state. The stable viewpoint values ​​are taken as the final simulation values ​​of the model.

[0104] This embodiment employs the opinion simulation evolution method based on an offline event incentive model. Combining this with the characteristic of social media platforms being influenced by the progress of offline events, a Hawkes process is used to describe the sequence of offline events, establishing an opinion evolution model that conforms to the characteristics of social media platforms. This model can predict the trend of user opinions in social media public opinion events, facilitating the control of public opinion dissemination on social media and contributing to the construction of a healthier and more harmonious online community space. Furthermore, this invention also captures the characteristic that extreme content on social media is restricted by platform norms. Based on a hyperbolic tangent function model, a social norm regulation factor is modeled, thereby capturing the characteristics of opinion evolution in user interactions on social media platforms from both online and offline perspectives. Compared with existing methods, this increases the dimension of influencing factors in the opinion evolution process, retains the core mechanism of opinion interaction, and achieves accurate fitting of the development process of social media public opinion events.

[0105] Example 2

[0106] Figure 3 This diagram illustrates the structure of a social media opinion evolution simulation device based on an offline event-driven model, as shown in this embodiment. Figure 3 As shown, the device includes: a public opinion event data processing and extraction module 301, an offline event excitation function module 302, a dynamic model parameter tuning module 303, and an opinion evolution simulation module 304.

[0107] The public opinion event data processing and extraction module 301 is used to acquire relevant social media data of public opinion events, extract the user interaction network structure from it, quantify the opinion tendencies in users' historical posts, and obtain the opinion values ​​of users at different time periods in the development of public opinion.

[0108] In an alternative approach, the public opinion event data processing and extraction module 301 is further used to: construct a network structure based on the social media data related to the acquired public opinion event, according to relationships such as comments, reposts, and follows.

[0109] In an alternative approach, the public opinion event data processing and extraction module 301 is further used to: encode the user's social media post text into a vector, and normalize it to obtain the opinion value by measuring the distance between it and a pre-selected vector of extreme opinions.

[0110] The offline event activation function module 302 is used to model the sequence of real events as a Hawkes point process based on public opinion event data and determine the relevant parameters of the event activation function.

[0111] In an alternative approach, the offline event activation function module 302 is further used to: model the real time series as a self-activating process, where each event increases the probability of another similar event occurring.

[0112] In an alternative approach, the offline event activation function module 302 is further used to: extract important time nodes from a series of related real events, represent them as time series point processes, and obtain the Hawkes form expression of the real events based on likelihood estimation.

[0113] The dynamic model parameter tuning module 303 is used to construct a viewpoint dynamic model based on offline event incentives, according to the model representing the user viewpoint interaction mechanism, the social norm regulation factor that regulates viewpoints to align with public opinion, and the Hawkes function of the real event; and to determine the parameters of the viewpoint dynamic model using the Monte Carlo method based on the user interaction network structure and user viewpoint values ​​in the public opinion event.

[0114] In an alternative approach, the dynamics model tuning module 303 is further used to: combine the social norm regulation factors and confidence threshold mechanisms present in the evolution mechanism of opinion dynamics to comprehensively form an opinion dynamics model based on offline event incentives.

[0115] In an alternative approach, the dynamic model tuning module 303 is further used to: tune parameters using a Monte Carlo method based on minimum average error, based on the network structure and user opinion values ​​in the public opinion event, to determine the optimal parameter combination.

[0116] The viewpoint evolution simulation module 304 is used to input the viewpoint dynamics model based on offline event incentives, according to the user's initial state and dynamic parameters obtained from real data, and iteratively calculate until the viewpoint values ​​of all users in the network converge to a stable state.

[0117] In an alternative approach, the viewpoint evolution simulation module 304 is further configured to: input the edge weight matrix and initial viewpoint values ​​into the viewpoint dynamics model, perform iterative calculations until the viewpoint values ​​of all users in the network converge to a stable state. The stabilized viewpoint values ​​are then taken as the final simulation values ​​of the model.

[0118] Example 3

[0119] This invention provides a non-volatile computer storage medium storing at least one executable instruction that can execute a social media opinion evolution simulation method based on an offline event-driven model in any of the above method embodiments.

[0120] Executable instructions can specifically be used to cause the processor to perform the following operations:

[0121] Obtain relevant social media data on public opinion events, extract the user interaction network structure from it, quantify the opinion tendencies in users' historical posts, and obtain the opinion values ​​of users at different time periods in the development of public opinion.

[0122] Based on public opinion event data, the sequence of real events is modeled as a Hawkes point process, and the relevant parameters of the event activation function are determined.

[0123] Based on the network structure, user opinion values, and Hawkes functions of real events extracted from the data, the parameters of the dynamic model based on offline event incentives are determined using the Monte Carlo method.

[0124] The user's initial state and dynamic parameters obtained from real data are input into the opinion dynamics model based on offline event incentives, and the calculation is iteratively performed until the opinion values ​​of all users in the network converge to a stable state.

[0125] Example 4

[0126] Figure 4 The diagram shows a structural schematic of an embodiment of the computing device of the present invention. The specific embodiments of the present invention do not limit the specific implementation of the computing device.

[0127] like Figure 4 As shown, the computing device may include:

[0128] Processor, Communications Interface, Memory, and Communications Bus.

[0129] The processor, communication interface, and memory communicate with each other via a communication bus. The communication interface is used to communicate with other network elements, such as clients or other servers. The processor executes programs, specifically the steps described in the embodiment of the social media opinion evolution simulation method based on an offline event-driven model.

[0130] Specifically, the program may include program code, which includes computer operation instructions.

[0131] The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The server may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0132] Memory is used to store programs. Memory may include high-speed RAM, and may also include non-volatile memory, such as at least one disk drive.

[0133] Specifically, the program can be used to cause the processor to perform the following operations:

[0134] Obtain relevant social media data on public opinion events, extract the user interaction network structure from it, quantify the opinion tendencies in users' historical posts, and obtain the opinion values ​​of users at different time periods in the development of public opinion.

[0135] The public opinion event is modeled as a Hawkes point process to obtain the first processing result;

[0136] The second processing result is obtained by processing the network structure and viewpoint training set data based on the Monte Carlo method;

[0137] Based on the network structure, initial viewpoint values, first processing results, and second processing results, predict the user viewpoint values ​​after reaching a stable state.

[0138] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, the embodiments of the present invention are not directed to any particular programming language. It should be understood that the content of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0139] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0140] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various inventive aspects, features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in fewer than all features of the single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0141] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0142] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, any of the claimed embodiments can be used in any combination.

[0143] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0144] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A simulation method for online opinion evolution based on an offline event-driven incentive model, characterized in that, include: Step S101: Obtain social media data about public opinion events, construct a user interaction network structure, quantify the opinion tendencies in users' historical posts, and obtain the opinion values ​​of users at different time periods during the development of public opinion. Step S102: Based on the public opinion event data, model the sequence of real events as a Hawkes point process to obtain the Hawkes function of the real events; Step S103: Based on the model representing the user opinion interaction mechanism, the social norm regulation factor that guides opinions to align with public opinion, and the Hawkes function of the real event, construct an opinion dynamics model based on offline event incentives; determine the parameters of the opinion dynamics model using the Monte Carlo method based on the user interaction network structure and user opinion values ​​in the public opinion event. The method for constructing the viewpoint dynamics model is as follows: First, based on the FJ model representing the interaction mechanism of user opinions and the social norm regulation factor that guides opinions towards the general public, a model of opinion evolution under the constraints of social norms is constructed as follows: Where, x i (t) represents the opinion value of user i, which changes with time t; K i Ω represents the social norm regulation coefficient for user i; α represents the neutral opinion value; α is the regulation parameter, the larger the α, the better the regulation effect; μ i h is the influence coefficient of user i's opinion. ij This represents the weight of the edge connecting user i and user j in the user interaction network structure; Therefore, the expression for the viewpoint dynamics model is: dx(t)=[(I n -Λ-K-ω)x(0)+Ku+Λh(x(t))]dt+ωdλ * (t) Among them, I n Let denot be the identity matrix, and Λ denote the diagonal matrix composed of sensitivity coefficients μ; Let y represent the feedback control vector of a social media platform for each user's opinion, and let y be the feedback control factor for user k. k (t)=-tanh(αx k (t)); K = diag(K1, K2, ..., K n ) represents the social norm regulation coefficient matrix; ω=diag(ω1,ω2,…,ω n ) represents the matrix showing the degree of influence of events occurring in the real space on different users, for any coefficient ω i ∈[0,1];λ * (t) represents the conditional strength function of an offline event in real space, i.e., the Hawkes function expression of a real event; dλ * (t) represents the conditional intensity variable within the time window [t, t+dt); h(x) represents h ij The edge weight matrix is ​​formed; Step S104: Input the initial user opinion values ​​obtained from the real data into the opinion dynamics model with determined parameters, and iterate until the opinion values ​​of all users in the user interaction network structure converge to a stable state, thus completing the online opinion evolution simulation.

2. The method as described in claim 1, characterized in that, The quantitative analysis of users' historical posts reveals their viewpoints at different points in time during the development of public opinion. Users' social media posts are encoded into vectors, and then normalized to obtain opinion values ​​by measuring the distance between them and pre-selected vectors of extreme opinions.

3. The method as described in claim 2, characterized in that, The vector distance between two texts is obtained by using the average cosine similarity between the word vectors in the text.

4. The method as described in claim 1, characterized in that, Step 102 involves modeling the real time series as a self-excitation point process, with the excitation function employing an exponential decay function. The Hawkes function of the real event is then expressed as: Where, λ base >0 represents the base strength, and t represents the current time. p Let t represent different times in the development of public opinion before t, with constants a, b > 0.

5. The method as described in claim 1, characterized in that, In step 103, during the process of determining the parameters of the opinion dynamics model using the Monte Carlo method, considering that user opinion interaction follows a confidence interval mechanism, the confidence interval is further used as a parameter tuning object to adjust the interaction weights h between interaction nodes in the user interaction network structure to be substituted into the opinion dynamics model. ij Perform a second zeroing check: If the difference in opinions between two interacting nodes is not within the confidence interval, no interaction will occur, and the interaction weight h between the two interacting nodes will be set. ij =0, and then substitute it into the viewpoint dynamics model.

6. The method as described in claim 1, characterized in that, The neutral opinion value Ω is the average of the maximum and minimum values ​​of each user's opinion in the user interaction network structure.

7. An online viewpoint evolution simulation device based on an offline event-driven model, used to execute the method described in any one of claims 1-6, characterized in that, The device includes: The public opinion event data processing and extraction module is used to acquire social media data about public opinion events, construct a user interaction network structure, quantify the opinion tendencies in users' historical posts, and obtain the opinion values ​​of users at different time periods in the development of public opinion. The offline event activation function module is used to model the sequence of real events as Hawke's point processes based on public opinion event data, and obtain the Hawke's function of the real events. The dynamics model parameter tuning module is used to construct a viewpoint dynamics model based on offline event incentives, according to the model representing the user viewpoint interaction mechanism, the social norm regulation factor that regulates viewpoints to align with public opinion, and the Hawkes function of the real event; and to determine the parameters of the viewpoint dynamics model using the Monte Carlo method based on the user interaction network structure and user viewpoint values ​​in the public opinion event. The opinion evolution simulation module is used to input the initial user opinion values ​​obtained from real data into the opinion dynamics model with defined parameters, and iteratively calculate until the opinion values ​​of all users in the user interaction network structure converge to a stable state, thus completing the online opinion evolution simulation.

8. A computing device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the online viewpoint evolution simulation method based on the offline event-driven model as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The storage medium stores at least one executable instruction that causes the processor to perform the operation corresponding to the online viewpoint evolution simulation method based on the offline event-driven model as described in any one of claims 1-6.

Images