LDA2Net Digging under the surface of COVID-19 scientific literature topics via a network-based approach.

During the COVID-19 pandemic, the scientific literature related to SARS-COV-2 has been growing dramatically. These literary items encompass a varied set of topics, ranging from vaccination to protective equipment efficacy as well as lockdown policy evaluations. As a result, the development of automatic methods that allow an in-depth exploration of this growing literature has become a relevant issue, both to identify the topical trends of COVID-related research and to zoom-in on its sub-themes. This work proposes a novel methodology, called LDA2Net, which combines topic modelling and network analysis, to investigate topics under their surface. More specifically, LDA2Net exploits the frequencies of consecutive words pairs (i.e. bigram) to build those network structures underlying the hidden topics extracted from large volumes of text by Latent Dirichlet Allocation (LDA). Results are promising and suggest that the topic model efficacy is magnified by the network-based representation. In particular, such enrichment is noticeable when it comes to displaying and exploring the topics at different levels of granularity.

The architecture of partisan debates: The online controversy on the no-deal Brexit.

We propose a framework to analyse partisan debates that involves extracting, classifying and exploring the latent argumentation structure and dynamics of online societal controversies. In this paper, the focus is placed on causal arguments, and the proposed framework is applied to the Twitter debate on the consequences of a hard Brexit scenario. Regular expressions based on causative verbs, structural topic modelling, and dynamic time warping techniques were used to identify partisan faction arguments, as well as their relations, and to infer agenda-setting dynamics. The results highlight that the arguments employed by partisan factions are mostly constructed around constellations of effect-classes based on polarised verb groups. These constellations show that the no-deal debate hinges on structurally balanced building blocks. Brexiteers focus more on arguments related to greenfield trading opportunities and increased autonomy, whereas Remainers argue more about what a no-deal Brexit could destroy, focusing on hard border issues, social tensions in Ireland and Scotland and other economy- and healthcare-related problems. More notably, inferred debate leadership dynamics show that, despite their different usage of terms and arguments, the two factions' argumentation dynamics are strongly intertwined. Moreover, the identified periods in which agenda-setting roles change are linked to major events, such as extensions, elections and the Yellowhammer plan leak, and to new issues that emerged in relation to these events.