Applications of information diffusion
https://www.sciencedirect.com/science/article/pii/S0370157316301600
Diffusion in networks is more than a theoretical subject. Modeling the information (fads, innovations, collective actions, viral memes, opinions and rumors) diffusion process is of outstanding interest for analyzing the information spreading patterns on social networks, stopping the transmission of viruses, as well as controlling the spread of rumors, etc. Models for static and dynamic networks have been developed. Most models concerning the adaptive process in information diffusion are used to characterizing the dynamic of opinion formation, such as the voter-like approach models and models concerning social segregation.
Thanks to the ever-increasing data collection of information spreading, applications of information spreading studies, ranging from network construction to opinion leader identification, benefit a lot for our human beings. The most famous applications include the multibillion dollar industry of instant messaging, blogging platforms like SNS, email networks, broadcasting, viral marketing, and crowdsourcing. Â
Another interesting facet of information diffusion is related to its application in predicting various social phenomena. Firstly, the approach of information diffusion has been used to predict the financial turbulence and regional economic trends. The financial market can be partially predicted since people tends to accept messages which are confirmed by social environment and then invest according to what they have learned. Secondly, in traffic systems, we sometimes intend to imitate how others choose routes to walk or drive if some unknown accidents happen, as such explicit behaviors can easily spread to the waiting queue. Finally, the online information in social networks, saying awareness, can somehow reflect the situation of epidemic spreading, hence can be applied to model and predict the coupling dynamics between information and epidemics from both micro- and macro-perspectives. In a word, the incorporation of various kinds of diffusion data can largely enhance our ability of understanding the underlying mechanisms of corresponding occurrences.
The science of information diffusion is just starting—despite impressive progresses, much remains to be understood. For further advances intuition alone is no longer enough and a multidisciplinary approach will surely bring powerful tools that innovative matchmakers may turn the immense potential into real life applications.