Yazar "Unver, Akin" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • [ X ]
    Öğe
    Analysis of Diplomatic Texts with LLMs: A Hybrid Approach in Computational Social Sciences
    (Ieee, 2025) Kurnaz, Ahmet; Unver, Akin
    This paper presents the practical application of big language models in text mining and topic modeling processes using texts shared online by the Ministry of Foreign Affairs of the Republic of Turkiye. The research describes an iterative and hybrid algorithm in which domain experts and large language models work together. The findings indicate that well-defined themes play a critical role in model success. It is emphasized that in social sciences, topic modeling processes should be handled with a tailor-made approach that includes the contextual interpretations of the researcher in the data evaluation processes, rather than being evaluated only through numerical outputs. In conclusion, the proposed algorithm offers a comprehensive and flexible analysis opportunity in social scientific research by generating topic labels suitable for qualitative or quantitative analysis of texts.
  • [ X ]
    Öğe
    Securitization of Disinformation in NATO's Lexicon: A Computational Text Analysis
    (Center Foreign Policy & Peace Research, 2022) Unver, Akin; Kurnaz, Ahmet
    Following the Russian meddling in the 2016 US elections, disinformation and fake news became popular terms to help generate domestic awareness against foreign information operations globally. Today, a large number of politicians, diplomats, and civil society leaders identify disinformation and fake news as primary problems in both domestic and foreign policy contexts. But how do security institutions define disinformation and fake news in foreign and security policies, and how do their securitization strategies change over years? Using computational methods, this article explores 238,452 tweets from official NATO and affiliated accounts, as well as more than 2,000 NATO texts, news statements, and publications since January 2014, presenting an unsupervised structural topic model (stm) analysis to investigate the main thematic and discursive contexts of these texts. The study finds that NATO's threat discourse and securitization strategies are heavily influenced by the US' political lexicon, and that the organization's word choice changes based on their likelihood of mobilizing alliance resources and cohesion. In addition, the study suggests that the recent disinformation agenda is, in fact, a continuity of NATO's long-standing Russiafocused securitization strategy and their attempt to mobilize the Baltic states and Poland in support of NATO's mission.