A study has found that estimating how contagious a tweet is from the first 50 retweets is the key to predicting whether a publication will be viral or not.
As social networks and online media continue to grow, so does the importance of understanding how they influence our thoughts and opinions, researchers said. Beihang University in China.
According to the study published in the journal PLOS ONE, it is considered a key objective to predict the spread of social contagion.
Although the models developed in the field of infectious diseases have been used to describe the dissemination of ideas, the studies have not used real data to estimate how infectious the information is.
The researchers used approximately one month of Twitter data, comprising more than 12 million tweets and more than 1.5 million retweets, and estimated the infectivity of each tweet based on the network dynamics of the first 50 retweets badociated with it. .
They incorporated the estimates of infectivity in a model with a decomposition constant that captures the gradual decrease in interest as online information ages.
Using real data and simulations, the researchers tested the capability of the infectivity-based model to predict the virality of retweet cascades.
They compared their performance with that of the standard community model, which incorporates other predictive factors, such as social reinforcement and capture effects that act to maintain tweet cascades within small communities of connected users.
The researchers found that for both the actual Twitter data and the simulated data, the infectivity model performed better than the community model, indicating that infectivity is a bigger driving force in determining whether a tweet becomes viral.
By combining the two models in a hybrid community, the infectivity model produced the most accurate predictions, highlighting the complexity of the interactive forces that determine the life and death of social network information.
"We propose a simulation model that uses Twitter data to show that infectivity, which reflects the intrinsic interest of a cascade of information, can substantially improve the predictability of viral cascades," the researchers said.
(This story has not been edited by the Business Standard staff and is generated automatically from a syndicated feed).