Search

Twitter is a social media platform where users can interact in three different ways: following, mentioning, or retweeting. Accordingly, one can define Twitter as a multilayer social network where each layer represents one of the three interaction mechanisms. First, we review the main findings of our previous work regarding two Twitter political conversations: the 2010 Venezuelan protest and the 2011 Spanish general elections. We found that the structure of the follower layer conditions the retweet layer, as having a low number of followers represents a constrain to effectively propagate information. The collapsed directed multiplex network does not present a rich-club ordering, as politicians presided large communities of regular users in the mention layer; while media accounts were the sources from which people retweeted information. However, when considering reciprocal interactions the rich-club ordering emerges, as elite accounts preferentially interacted among themselves and largely ignored the crowd. Finally, we explore the main relationships between the community structure of the three layers. At the follower level users cluster in large and dense communities holding various hubs, that break into smaller and more segregated ones in the mention and retweet layers. Hence, we argue that to fully understand Twitter we have to analyze it as a multilayer social network, evaluating the three types of interactions. [ABSTRACT FROM AUTHOR]

Anticipating price crises in the market of agri-commodities is critical to guarantee both the sustainability of the food system and to ensure food security. However, this is not an easy task, since the problem implies analyzing small and very volatile time series, which are highly influenced by external factors. In this paper, we show that suitable reservoir computing algorithms can be developed that outperform traditional approaches, by reducing the Mean Absolute Error and, more importantly, increasing the Market Direction Accuracy. For this purpose, the applicability of five variants of such method to forecast this market is explored, and their performance evaluated by comparing the results with those obtained with the standard LSTM and SARIMA benchmarks. We conclude that decomposing the time series and modeling each component with a separate RC is essential to successfully anticipate price trends, and that this method works even in the complex changing temporal scenario of the Covid-19 pandemic, when part of the data were collected. • We tackle current food loss problem. • Explore predictability in agri-food market. • Successfully anticipate sudden price crisis. • Use the recent ML Reservoir Computing method. • We outperform traditional benchmark models: SARIMA and LSTM. [ABSTRACT FROM AUTHOR]