Information diffusion on social media is a quite complex phenomenon to be analyzed, since these are free and easily accessible to anyone who has an Internet connection and a proper device. There are several mathematical approaches to carry out this kind of analysis: one of these consists of using epidemiological models based on ordinary differential equations [2, 3, 4]. However, describing only the evolution of the phenomenon is not sufficient, but it is even required to predict its evolution.

The main aim of this talk is to highlight how, using a proper parameter estimation strategy and an adequate dataset, built using real data, it is possible to obtain the desired predictions [1], as showed by numerical tests realized studying real news spread on the social network X (Twitter) during the period 2020-2022.

References

1. Castiello, M.; Conte, D.; Iscaro, S. Using Epidemiological Models to Predict the Spread of Information on Twitter. Algorithms 2023, 16, 391. https://doi.org/10.3390/a16080391

2. D’Ambrosio, R., Giordano, G., Mottola, S., Paternoster, B., Stiffness analysis to predict the spread out of fake information. Future Internet 13(9) (2021), 222.

3. Maleki, M., Mead, E., Arani, M., Agarwal, N., Using an epidemiological model to study the spread of misinformation during the Black Lives Matter Movement. arXiv:2103.12191 (2021).

4. Muhlmeyer, M., Agarwal, S., Information spread in a social media age. Modelling and Control, CRC Press (2021), Taylor and Frencis Group: Boca Raton, London, New York.

[link to pdf] [back to Numdiff-17]