Media narratives and the social perception of violence

Abstract

The model proposed in this article faces the necessity to define a computational framework designed to analyze how media narratives
shape the perception and experience of violence. The starting point of the analysis is a corpus derived from The Guardian, which
serves as a foundation for broader extensions to include more diverse media sources. The aim is to investigate how perpetrators of
violence may feel emboldened by societal indifference or institutional inertia, while victims may experience a deepening sense of
isolation and social alienation. The process begins with the preprocessing of textual data to construct a clean and structured corpus,
enabling the application of advanced computational methods. Through document embedding and dimensionality reduction
techniques such as t-SNE, the model identifies latent patterns and relationships within the data, making them visually accessible.
This is complemented by keyword extraction and ontology mapping, which work together to highlight critical terms and organize
them into conceptual frameworks. These steps allow for a deeper understanding of how narratives around violence are constructed
and disseminated. The integration of topic modelling and sentiment analysis further enriches this understanding by uncovering
dominant themes and emotional undertones, shedding light on how violence is framed in public discourse. These insights are
translated into network representations, revealing the intricate interconnections between entities, events, and sentiments. The
resulting visualizations and statistical outputs provide an interactive lens through which researchers can explore the dynamics of
media reception and its psychological and social consequences. By linking computational insights to the human experience of trauma, this model facilitates the identification of patterns that illuminate the ways in which media narratives influence both individual and collective responses to violence. The iterative refinement of this methodology seeks to bridge computational analysis with the socio emotional dimensions of trauma, ultimately offering a powerful tool to interrogate the societal structures that perpetuate harm or, conversely, foster healing and resilience.