Information diffusion prediction method and device

Abstract

The embodiment of the invention provides an information diffusion prediction method and device. The information diffusion prediction method comprises the following steps: acquiring a current user transmitting preset information, user characteristics of the current user, a previous user and user characteristics of the previous user; mapping the previous user and the user characteristics of the previous user into a potential vector space to obtain a first user junction and a first user characteristic vector, and mapping the current user and the user characteristics of the current user into the potential vector space to obtain a second user junction and a second user characteristic vector; establishing a potential vector space model according to the first user junction, the first user characteristic vector, the second user junction and the second user characteristic vector; learning the potential vector space model to obtain a kernel function formula relevant to the first user junction, the first user characteristic vector, the second user junction and the second user characteristic vector; predicting a transmission trend indicating that the second user junction transmits preset information to at least one adjacent junction in other adjacent junctions according to the kernel function formula.

Description

technical field [0001] The invention relates to the field of information dissemination, in particular to a method and device for information diffusion prediction. Background technique [0002] In recent years, relevant research in the field of social network analysis has been paid more and more attention, and information dissemination has become one of the research focuses. At present, with the deepening of research, research in the field of information dissemination has been gradually divided into four categories: community discovery, social influence prediction, information source detection and link prediction. There is also a subdivided research field in link prediction, that is, propagation prediction, whose goal is to predict which groups of people the information will eventually affect in social networks. [0003] Currently, there are two types of propagation prediction models, similarity models and probability models. The prediction of the similarity model is based ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the embodiments of the present invention is to provide a method and device for predicting information diffusion, so as to solve the problem that a large amount of user information is required for prediction in current information dissemination

Images