Mapping Opinion dynamics and Recommender systems research

These maps have been produced for the state of the art of Chavalarias, D., Bouchaud, P., Panahi, M., 2024. Can a Single Line of Code Change Society? The Systemic Risks of Optimizing Engagement in Recommender Systems on Global Information Flow, Opinion Dynamics and Social Structures. JASSS 27, 9.

Key domain expressions (nodes in the map) were extracted by means of text mining on the title and abstracts of the selected documents. The semantic proximity between these expressions was calculated as the metric confidence, i.e., the maximum of the two conditional probabilities of having one expression knowing the other in the same publication. The resulting graph was visualized using Gephi.

Opinion dynamics landscape

The map of the opinion dynamics litterature has been made with GarganText from 1,872 publications extracted from the Web of Sciences. The WoS query: "Opinion dynamics" [2023-06-10]. The source of this maps is available on the archive (Bouchaud, Chavalarias, & Maziyar 2023).

Recommender systems and social media literature

The map of the Recommender systems and social media literature has been made with GarganText from 1,267 publications extracted from the Web of Sciences. Query in the WoS on 11/06/2023: ("recommender systems" OR "content recommendation") AND (Twitter OR facebook OR "social network" OR "online social media").

How to read these maps?

This map identifies the main vocabulary of the opinion dynamics literature and establishes the relationships that have been established in its papers via their confidence similarity (max of the two conditional probabilities of having one term in the meta-data of a paper knowing the presence of the other).

This map was done with Gargantext on March 10 2023