Martina Katalin Szabó

Martina Katalin Szabó
Download CV Research Fellow (TK Recens)
Research Interests

As a linguist and NLP researcher, my work bridges the fields of sociolinguistics, computational linguistics, and mental health. A central focus of my research is understanding how linguistic usage patterns reflect individuals' social and cognitive dimensions.

Currently, I am engaged in two major research areas. The first focuses on the intersection of language and mental health, where I analyze linguistic markers associated with various mental health conditions, such as stroke, dysarthria, schizophrenia, bipolar disorder, and dementia, through computational methods. This research investigates how language use reflects cognitive states and contributes to understanding the linguistic dimensions of these disorders, including the effects of depression and anxiety on language production. A key objective of this work is to enable the prediction of certain psychological states using NLP methods and tools based on linguistic characteristics.

The second area of my research examines the relationship between sociological status and linguistic behavior in online environments. In this project, I utilize large-scale Facebook data and NLP analysis techniques to investigate how linguistic traits correlate with users' socioeconomic status. Additionally, I explore how these linguistic factors position users within a model, providing insights into social strata beyond mere prediction. By uncovering these correlations, the study can enhance our understanding of the interrelations between social inequalities and digital literacy, providing insights into the relationship between language and social dynamics.

Selected Publications
  • Ring, O., Szabó, M. K., Guba, Cs., Váradi, B., Üveges, I. (2023). Approaches to Sentiment Analysis of Hungarian Political News at the Sentence Level. Language Resources and Evaluation. Language Resources and Evaluation, 1-29. Ranking: Q1, Impact factor: 2.7.
  • Szabó, M. K., Vincze, V. and Bibok, K. (2022). "Thank you for the terrific party!" – An Analysis of Hungarian Negative Emotive Words. Corpus Linguistics and Linguistic Theory, 2022. Open access: https://www.degruyter.com/document/doi/10.1515/cllt-2022-0013/html. Ranking: Q1, 5-year Journal Impact Factor: 1.9.
  • Egas-López, J. V., Balogh, R., Imre, N., Hoffmann, I., Szabó, M. K., Tóth, L., ... & Gosztolya, G. (2022). Automatic Screening of Mild Cognitive Impairment and Alzheimer’s Disease by Means of Posterior-Thresholding Hesitation Representation. Computer Speech & Language, 75, 101377. Ranking: Q1, Journal Impact Factor: 4.3.
  • Vincze, V., Szabó, M. K., Hoffmann, I., Tóth, L., Pákáski, M., Kálmán, J., & Gosztolya, G. (2022). Linguistic Parameters of Spontaneous Speech for Identifying Mild Cognitive Impairment and Alzheimer Disease. Computational Linguistics, 48(1), 119–153. Ranking: Q1, Journal Impact Factor: 9.3.
  • Gulyás, A., Szabó, M. K., Ring, O., Kiss, L., Boros, I. (2021). Networks of the Political Elite and Political Agenda Topics: Creation and Analysis of Historical Corpora Using NLP and SNA Methods. In Rudas, T., Péli, G. (Eds.) Pathways Between Social Science and Computational Social Science. Computational Social Sciences. Springer, Cham. DOI: https://doi.org/10.1007/978-3-030-54936-7_9, 197–214. Ranking: D1.
  • Szabó, M. K., Ring, O., Nagy, B., Kiss, L., Koltai, J., Berend, G., Vidács, L., Gulyás, A. & Kmetty, Z. (2020). Exploring the Dynamic Changes of Key Concepts of the Hungarian Socialist Era with Natural Language Processing Methods. In Historical Methods: A Journal of Quantitative and Interdisciplinary History, 54(1), 1–13. DOI: 10.1080/01615440.2020.1823289, https://www.tandfonline.com/doi/full/10.1080/01615440.2020.1823289. Ranking: Q1, Journal Impact Factor: 1.3.
Research Projects
  • 2024 September - present:  Volunteer researcher in the MTA–TK Lendület “Momentum” Digital Social Science Research Group for Social Stratification Title of the research: Social Structure and Inequalities Through the Lens of Digital Data. Principal Investigator: Júlia Koltai. Host Institute: Computational Social Science - Research Center for Educational and Network Studies (CSS-RECENS) of the HAS Centre for Social Sciences, Hungary.
  • 2024.04.01 – Present: Artificial Intelligence National Laboratory (MILAB) subproject. Title of the research: ComMent: Computational Methods for Mental Health - For the deeper understanding of the connections among stroke, communication challenges, and psychosocial factors. Principal Investigator: Martina Katalin Szabó. Host Institute: Computational Social Science - Research Center for Educational and Network Studies (CSS-RECENS) of the HAS Centre for Social Sciences, Hungary.
  • 2020 - 2024.09.30: OTKA Postdoctoral Fellowship (NKFI / OTKA). Proposal number: 132312; Title of the proposal: A corpus-based computational analysis of Hungarian negative emotive elements from the viewpoint of semantic changes and mental disorders. Principal Investigator: Martina Katalin Szabó. Host Institute: University of Szeged, Department of Software Engineering, Hungary.
  • 2019 - 2024: Hungarian Research Fund (NKFIH / OTKA). Proposal number: 131826; Title of the proposal: Social history analysis of the press between 1945 and 1989 with the methods of natural language processing (NLP). Principal Investigator: Júlia Koltai. Role: Investigator. Host Institute: Computational Social Science - Research Center for Educational and Network Studies (CSS-RECENS) of the HAS Centre for Social Sciences, Hungary.
  • 2016-2018, 2019 - 2022: EVILTONGUE (ERC), ERC Consolidator Grant 2014. Proposal number: 648693; Title of the proposal: No Sword Bites So Fiercely as an Evil Tongue? Gossip Wrecks Reputation, but Enhances Cooperation. Principal Investigator: Károly Takács. Role: Investigator.