Search

CN-122022785-A - Tibetan Chinese rumor propagation model based on coupling state and construction method thereof

CN122022785ACN 122022785 ACN122022785 ACN 122022785ACN-122022785-A

Abstract

A Tibetan rumor propagation model based on coupling state is represented by a user model in a susceptibility state as shown in a formula (1) User model for Chinese contact state Tibetan language contact state user model User model of coupling state User model for recovery state Composition is prepared. The method for constructing the model comprises the steps of dividing a social network user state model, propagating rumors, determining discrete step sizes, determining model balance points, determining stability conditions of the propagation model balance points and constructing the model. The method solves the technical problem that in the prior art, the cross-language coupling diffusion is difficult to accurately describe due to mutually independent modeling of different language propagation processes, can more clearly describe the complex propagation process of rumors propagation caused by cross-language contact, bilingual propagation selection and a coupling conversion mechanism, improves the depicting capability of propagation strength and evolution trend of the Tibetan-Han bilingual rumors, and is suitable for situation assessment and early warning of Tibetan-Han social network rumors propagation.

Inventors

  • WANG ZHENGBIN
  • WANG XIAOMING
  • LIN YAGUANG

Assignees

  • 青海师范大学

Dates

Publication Date
20260512
Application Date
20260130

Claims (4)

  1. 1.A Tibetan rumor propagation model based on coupling state is characterized in that the model User model from susceptible state as shown in (1) User model for Chinese contact state Tibetan language contact state user model User model of coupling state User model for recovery state The composition is as follows: (1) Wherein, the Representing a discrete step size of the device, Indicating the total user scale at the last moment; 、 、 、 、 The user model is a susceptibility state user model, a Chinese contact state user model, a Tibetan contact state user model, a coupling state user model and a recovery state user model at the last moment respectively, and parameters are as follows For a susceptible state user access rate, Off-network rate for the user; for the chinese contact rumor state coefficient, For the Tibetan language contact rumor state coefficients, In order to propagate rumor coefficients in chinese, In order to propagate rumor coefficients in Tibetan language, Is the coefficient of the coupling state of the Chinese language, Is the Tibetan language coupling state coefficient, Is a coefficient of the state of Chinese restoration, Is the state coefficient of the Tibetan language recovery, State coefficients are recovered for Tibetan chinese language.
  2. 2. The coupling-state-based Tibetan rumor propagation model of claim 1, wherein in formula (1), the coefficients 、 、 、 、 、 、 、 、 、 The value range of (1, 0), said The value range is [5,20], The value range is [0.01,1.0].
  3. 3. The coupling-state-based Tibetan rumor propagation model according to claim 1 or 2, wherein in formula (1), the The value of the water-based paint is 0.28, The value of the water-based paint is 0.22, The value of the water-based paint is 0.88, The value of the water-based paint is 0.80, The value of the water-based paint is 0.14, The value of the water-based paint is 0.10, The value of the water-soluble polymer is 0.05, The value of the product is 0.07, The value of the water-based paint is 0.08, The value of the water-soluble fluorescent powder is 0.03, The value of the water-based paint is 10, The value is 0.5.
  4. 4. A method for constructing a coupling-state-based Tibetan rumor propagation model as claimed in claim 1, comprising the steps of: (1) Partitioning social network user state model At discrete moments Based on the monitoring data of the social network platform, dividing the user into a user model with a susceptibility state at the last moment User model for Chinese contact state Tibetan language contact state user model User model of coupling state User model for recovery state ; (2) Rumor spreading process User model of susceptibility state at last moment At the rate of network access Entering an online social network, and enabling user models in all states to be off-network coefficients User model of Chinese contact state at last moment when rumors are propagated Spreading rumor coefficients in chinese Participating in propagation, coupled state user model at last moment Synergistic effect on transmission in discrete steps In each of the following transfer amounts, the time of each of which is discrete Is determined by each state user model of the system and counts the state update at the next moment, and the susceptibility state user model at the last moment In contact with rumor state coefficient Under the action of (a) at an instantaneous flow rate User model for turning into Chinese contact state at last moment ; Tibetan language contact state user model at last moment Propagation of rumor coefficients in Tibetan Participating in propagation, coupled state user model at last moment Synergistic effect on transmission in discrete steps In, the user model of the susceptibility state at the last moment Tibetan language contact rumor state coefficient Under the action of (a) at an instantaneous flow rate Tibetan language contact state user model for switching to last moment ; During coupling and recovery, the user model of Chinese contact state at the last moment By Chinese coupling state coefficients User model for coupling state of last moment Tibetan language contact state user model at last moment Tibetan language coupling state coefficient User model for coupling state of last moment Chinese contact state user model at last moment By Chinese recovery state coefficients User model for switching to recovery state at last moment Tibetan language contact state user model at last moment Tibetan language recovery state coefficient User model for switching to recovery state at last moment User model of coupling state at last moment Recovering state coefficient in Tibetan language Switching to a user model in a recovery state at the last moment; (3) Determining discrete steps Determining discrete step length according to data sampling interval method ; (4) Determining model equilibrium points Based on equation (1), when each step increment of each state is 0, the system reaches equilibrium, and a balance point equation set is obtained as follows: Wherein, the 、 、 、 、 Respectively taking values of applications under balance points, and constructing rumor-free propagation balance points by a balance equation set Equilibrium point with rumor propagation : 1) Construction of rumor-free propagation equilibrium Point Building rumor-free propagation equilibrium points as follows : 2) Construction of rumor propagation equilibrium Point Building rumor propagation equilibrium points as follows : , (5) Determining propagation model equilibrium point stability conditions 1) Constructing propagation model equilibrium points Linearization operator at Constructing propagation model equilibrium points as follows Linearization operator at : , Wherein, the 、 Is a state vector of 5 dimensions and, Representing a transpose; 2) Constructing propagation model equilibrium points Characteristic value at Constructing propagation model equilibrium points as follows Characteristic value at : Wherein, the The eigenvalues of the matrix are represented, ; 3) Determining rumor-free propagation equilibrium point Stability conditions of (2) When (when) Balance point at the time Global asymptotic stabilization; 4) Determining rumor propagation balance point Stability conditions of (2) When (when) , And (2) and Balance point at the time Local asymptotic stabilization; (6) Building a model And (3) constructing a Tibetan rumor propagation model shown in the formula (1) according to the state division, the parameters and the stability conditions determined in the steps (1) - (5).

Description

Tibetan Chinese rumor propagation model based on coupling state and construction method thereof Technical Field The invention belongs to the technical field of computer application, and particularly relates to information transmission. Background Rumors in an online social network are fast in propagation, strong in cross-layer diffusion and the like. In multilingual environments, rumors are often cross-propagated between different language rings by bilingual users, forming cross-lingual coupling diffusion. The existing research has a certain limitation in the aspect of multilingual rumor propagation modeling, for example, multilingual coupling user groups are not explicitly defined, or different language propagation processes are regarded as mutually independent submodels, so that it is difficult to describe a co-propagation mechanism and cross-language association propagation rules. In a Tibetan social network scene, users often contact information sources of two languages at the same time, and differences of propagation intensity and propagation paths can be generated in translingual transshipment, transshipment and synonymous rewriting processes of rumor content, so that a single language model is difficult to reflect a real diffusion link. In particular, bilingual users act as "bridging nodes" between different ring layers, and their propagation choices amplify or suppress cross-language diffusion, and enable the propagation processes of the two languages to exhibit coupling evolution characteristics. The existing method depends on continuous time or static statistics assumption, and is difficult to align with discrete time characteristics of platform sampling data, so that situation assessment, trend prediction and early warning decision-making are not facilitated. Therefore, it is necessary to construct a Tibetan rumor propagation model oriented to discrete time update and capable of explicitly describing the coupling state and cross-language co-propagation mechanism. Disclosure of Invention The technical problem to be solved by the invention is to overcome the defects of the prior art, and provide the Tibetan rumor propagation model based on the coupling state, which can characterize a cross-language coupling propagation mechanism. Another technical problem to be solved by the present invention is to provide a method for constructing the rumor propagation model. The technical scheme adopted for solving the technical problems is as follows: tibetan rumor propagation model based on coupling state, and model thereof User model from susceptible state as shown in (1)User model for Chinese contact stateTibetan language contact state user modelUser model of coupling stateUser model for recovery stateThe composition is as follows: (1) Wherein, the Representing a discrete step size of the device,Indicating the total user scale at the last moment;、、、、 The user model is a susceptibility state user model, a Chinese contact state user model, a Tibetan contact state user model, a coupling state user model and a recovery state user model at the last moment respectively, and parameters are as follows For a susceptible state user access rate,Off-network rate for the user; for the chinese contact rumor state coefficient, For the Tibetan language contact rumor state coefficients,In order to propagate rumor coefficients in chinese,In order to propagate rumor coefficients in Tibetan language,Is the coefficient of the coupling state of the Chinese language,Is the Tibetan language coupling state coefficient,Is a coefficient of the state of Chinese restoration,Is the state coefficient of the Tibetan language recovery,State coefficients are recovered for Tibetan chinese language. In formula (1), the coefficients、、、、、、、、、The value range of (1, 0), saidThe value range is [5,20],The value range is [0.01,1.0]. In the formula (1), theThe value is optimally set to be 0.28,The value is optimally set to be 0.22,The value is optimally set to be 0.88,The value is optimally set to be 0.80,The value is optimally set to be 0.14,The value is optimally set to be 0.10,The value is optimally set to be 0.05,The value is optimally set to be 0.07,The value is optimally set to be 0.08,The value is optimally set to be 0.03,The value is optimally 10, and the total number of the components is 10,The optimal value is 0.5. The above modelThe construction method of the method comprises the following steps: (1) Partitioning social network user state model At discrete momentsBased on detection data of the social network platform, dividing users into user models in susceptibility states at the last momentUser model for Chinese contact stateTibetan language contact state user modelUser model of coupling stateUser model for recovery state。 (2) Rumor spreading process User model of susceptibility state at last momentAt the rate of network accessEntering an online social network, and enabling user models in all states to be off-network coefficientsUser mode