Mapping the evolution of cross-Strait relations via global news big data (2014–2023): An analysis integrating GDELT and machine learning

This study investigates global news media representations of cross-Strait relations from 2014 to 2023 using the GDELT database, framed within the mediatization of politics. Combining large-scale event data and computational text analysis, it offers a multi-level analysis of representational patterns, structural inequalities, and framing dynamics. Six indicators track longitudinal trends including four continuous indices and two event type distributions, show attention spikes during major political events and a discursive shift toward negative, conflict-focused coverage, even as low-intensity communicative events dominate. Structurally, a source-domain analysis of 50 high-impact outlets employs a six-dimensional Deviation Index to evaluate differences in visibility and event production preferences. Results reveal a concentrated discourse, with Western and Taiwanese media occupying more central positions in agenda-setting processes, with Mainland Chinese outlets appearing comparatively less visible. Textually, topic modeling uncovers five key frames, reflecting a discursive thematic evolution from event-driven, low-politics coverage to high-politics narratives emphasizing conflicts, ideological divides, and great-power rivalry. Complementary sentiment analysis of news headlines using large language model-based tools indicates the persistent dominance of negative actor-sentiment framing in global coverage. Overall, the study underscores an increasingly securitized and asymmetrical pattern of global media representation of cross-Strait relations. Theoretically, the study extends the applicability of mediatization of politics framework to the analysis of cross-Strait communication. Methodologically, it illustrates integrating large-scale event data with machine-learning techniques to examine international news framing in a highly politicized geopolitical context.