In this paper, we present the outcome of a structural linguistic analysis performed on a referentially grounded FrameNet dataset. In this dataset, multiple Dutch events are referenced by multiple co-referential Dutch news texts. Mentions in those documents are annotated with respect to their referential grounding (i.e., links to structured Wikidata), and their conceptual representation (i.e., frames). Provided with each document’s temporal reporting distance, we selected documents for two events - the Utrecht shooting and MH17 - and performed an analysis in which we tracked the events’ participants over time in both their focalization (number of mentions) and their framing (distribution of frame element labels). This way, we use the carefully collected and annotated data to schematize shifts in focalization and perspectivization of the participants as a result of the constantly developing narrative surrounding the events. This novel type of linguistic research involves reference to the real-world referents and takes into account storytelling in news streams.

In this work we introduce TimeFrame, an online platform to easily query and visualize events and participants extracted from document collections in Italian following a frame-based approach. The system allows users to select one or more events (frames) or event categories and to display their occurrences on a timeline. Different query types, from coarse to fine-grained, are available through the interface, enabling a time-bound analysis of large historical corpora. We present three use cases based on the full archive of news published in 1948 by the newspaper “Corriere della Sera”. We show that different crucial events can be explored, providing interesting insights into the narratives around such events, the main participants and their points of view.

We present an experiment on classifying news frames in a language unseen by the learner, using zero-shot cross-lingual transfer learning. We used two pre-trained multilingual Transformer Encoder neural network models and tested with four specific news frames, investigating two approaches to the resulting multi-label task: Binary Relevance (treating each frame independently) and Label Power-set (predicting each possible combination of frames). We train our classifiers on an available annotated multilingual migration news dataset and test on an unseen Slovene language migration news corpus, first evaluating performance and then using the classifiers to analyse how media framed the news during the periods of Syria and Ukraine conflict-related migrations.

We introduce a large corpus of comments extracted from an Italian online incel (‘involuntary incelibate’) forum, a community of men who build a collective identity and anti-feminist ideology centered around their inability to find a sexual or romantic partner and who frequently use explicitly misogynistic language. Our corpus consists of 2.4K comments that have been manually collected, analyzed and annotated with topic labels, and a further 32K threads (300K comments) that have been automatically scraped and automatically annotated with FrameNet annotations. We show how large-scale frame semantic analysis can shed a light on what is discussed in the community, and introduce incel topic classification as a new NLP task and benchmark.

Current approaches to computational metaphor processing typically incorporate static representations of metaphor. We aim to show that this limits the coverage of such systems. We take insights from dynamic metaphor theory and discuss how existing computational models of metaphor might benefit from representing the dynamics of metaphor when applied to the analysis of conflicting discourse. We propose that a frame-based approach to metaphor representation based on the model of YinYang Dynamics of Metaphoricity (YYDM) would pave the way to more comprehensive modeling of metaphor. In particular, the metaphoricity cues of the YYDM model could be used to address the task of dynamic metaphor identification. Frame-based modeling of dynamic metaphor would facilitate the computational analysis of perspectives in conflicting discourse, with potential applications in analyzing political discourse.