Abstract: Considering traditional research on social-ecological crises, new social media analysis, particularly Twitter data, contributes with supplementary exploration techniques. In this article, we argue that a social media approach to social-ecological crises can offer an actor-centered meaningful perspective on social facts, a depiction of the general dynamics of meaning making that takes place among actors, and a systemic view of actors' communication before, during and after the crisis. On the basis of a multi-technique approach to Twitter data (TF-IDF, hierarchical clustering, egocentric networks and principal component analysis) applied to a red tide crisis on Chiloe Island, Chile, in 2016, the most significant red tide in South America ever, we offer a view on the boundaries and dynamics of meaning making in a social-ecological crisis. We conclude that this dynamics shows a permanent reflexive work on elucidating the causes and effects of the crisis that develops according to actors' commitments, the sequence of events, and political conveniences. In this vein, social media analysis does not replace good qualitative research, it rather opens up supplementary possibilities for capturing meanings from the past that cannot be retrieved otherwise. This is particularly relevant for studying social-ecological crises and supporting collective learning processes that point towards increased resilience capacities and more sustainable trajectories in affected communities.

Abstract: Constitutional processes are a cornerstone of modern democracies. Whether revolutionary or institutionally organized, they establish the core values of social order and determine the institutional architecture that governs social life. Constitutional processes are themselves evolutionary practices of mutual learning in which actors, regardless of their initial political positions, continuously interact with each other, demonstrating differences and making alliances regarding different topics. In this article, we develop Tree Augmented Naive Bayes (TAN) classifiers to model the behavior of constituent agents. According to the nature of the constituent dynamics, weights are learned by the model from the data using an evolution strategy to obtain a good classification performance. For our analysis, we used the constituent agents' communications on Twitter during the installation period of the Constitutional Convention (July-October 2021). In order to differentiate political positions (left, center, right), we applied the developed algorithm to obtain the scores of 882 ballots cast in the first stage of the convention (4 July to 29 September 2021). Then, we used k-means to identify three clusters containing right-wing, center, and left-wing positions. Experimental results obtained using the three constructed datasets showed that using alternative weight values in the TAN construction procedure, inferred by an evolution strategy, yielded improvements in the classification accuracy measured in the test sets compared to the results of the TAN constructed with conditional mutual information, as well as other Bayesian network classifier construction approaches. Additionally, our results may help us to better understand political behavior in constitutional processes and to improve the accuracy of TAN classifiers applied to social, real-world data.