News

The Text Mining and Artificial Intelligence Training Program (foundation course) was successfully held between the 11th and 12th of November, 2021 with 25 participants. The program introduced the theoretical and practical foundations of the social science applications of text mining and artificial intelligence, and provided participants with the opportunity to complete a mini-project.

A text mining and artificial intelligence training program (introductory course) will be introducing the theoretical and practical foundations of the social science applications of text mining and artificial intelligence and provide the participants with the opportunity to complete a mini-project. The course will be held from 11 to 12 November 2021 in a 3×1.5 hour frame (9: 00-10: 30, 10: 45-12: 15 and 13: 15-14: 45) in the HTK K013-14 meeting room, performed by Miklós Sebők, Ákos Máté, Orsolya Ring and Péter István Járay.

The 4th International and Interdisciplinary Conference on the Quantitative and Computational Analysis of Textual Data (COMPTEXT) will be held in Dublin, Ireland, on 6–7 May 2022. COMPTEXT is an international community of quantitative text analysis and computational social science scholars in political science, international relations and beyond. COMPTEXT 2022 in Dublin follows in the footsteps of previous conferences in Budapest (2018), Tokyo (2019) and Innsbruck (online, 2020). COMPTEXT conferences offer ample opportunities to network with computational scholars, exchange technological knowledge of computational methods, and obtain useful feedback on ongoing research. For COMPTEXT 2022 in Dublin we are seeking paper submissions that We accept both substantive and methodological papers for presentation: substantive papers may be on any studies in social sciences or humanities that utilize computational methods; methodological papers may describe new computational methods, tools and approaches. In keeping with our tradition, ahead of the conference a series of methods training tutorials will be held for registered participants. Courses will be offered for both beginner and advanced level participants. Submission of Paper Abstracts: Abstracts of 200 words and a number of substantive and methods-related keywords, should be submitted by Friday 28 January 2022. Notifications of acceptance will be sent by 18 February 2022. The registration deadline is 15 March 2022. Please submit your paper at https://forms.gle/bvq8rh86awdtGNhLA Please be advised that a modest conference fee will be charged for participants with accepted papers. The COMPTEXT 2022 Organising Committee consists of: Equality, Diversion, and Inclusion: COMPTEXT is committed to creating an inclusive conference where diversity is celebrated, and everyone is afforded equality of opportunity. We welcome applications from everyone, including those who identify with any of the protected characteristics that are set out in UCD’s Equality, Diversity and Inclusion policy (https://www.ucd.ie/equality/). We especially encourage scholars from traditionally under-represented groups, female scholars, and early-career researchers to apply. Conference Format and Health Regulations: Note that in line with current Irish regulations, this will be a “green pass” event. Only those with a valid EU Digital COVID Certificate, NHS COVID Pass, or alternative proof of vaccination will be able to access the conference venue. If the regulatory regime changes we will update the requirement accordingly – including moving the event online should the in-person event become unfeasible. For more information, please visit our website: http://www.comptextconference.org/ Questions related to COMPTEXT Dublin 2022 should be directed to COMPTEXT22@ucd.ie Best regards, The Organizers

The online series of events of the TK MILAB Speaker Series of the National Laboratory of Artificial Intelligence of the Research Center for Social Sciences continues. The following presentation will be on October 12, 2021, from 10:00 am. “The use of the huBert model in emotion analysis – an attempt to classify low-resource text on Hungarian-language news corpora” by Orsolya Ring and Márk Kis György. The presentation will be in Hungarian, online, on the Zoom application. Participation in the event is subject to registration, available at the following link (https://cutt.ly/oENwpRv).

10 June, 2021 – Policy Agendas in Autocracy, and Hybrid Regimes (Eds. Miklós Sebők, Zsolt Boda) book is published Information about the upcoming book launch will be published later, please follow our website. The published book is available here.

29 April, 2021 – Speech by Orsolya Ring at conference organised by MILAB and the Digital Wellbeing Program Orsolya Ring gave a talk entitled ‘Measuring Customer Satisfaction through the Evaluation of Texts’ at the online conference jointly organised by MILAB and the Digital Wellbeing Program (Digitális Jólét Program). The online conference examined the effects the recent boom in language technology has had on customer services. More information about the conference is available on the event’s website (in Hungarian).

14 April, 2021 – New publication: Article by Miklós Sebők, Kristin Makszin and Jasper Simons in East European Politics Miklós Sebők, Kristin Makszin and Jasper Simons’ new article entitled ’Mission adapted: the hidden role of governors in shaping central bank operating missions in Hungary’ has been published in the East European Politics journal. ABSTRACT: ’Despite the diffusion of the paradigm of central bank independence, there is still meaningful variation in the operating missions of central banks both across countries and over time. Through a detailed qualitative case study, this article develops the concept of the operating mission of the central bank and applies it to the case of the Hungarian National Bank (MNB) to provide a more complete understanding of mission shift. Our findings demonstrate the critical role of policy agency, as the central bank governors moulded the operating mission of the central bank, even in the face of dominant international norms.’ The full article is available here. (OpenAccess)

14 April, 2021 – The CLARIN project ends with fruitful results poltextLAB’s CLARIN project came to a successful end. The annotated corpus consisting of 3086 parliamentary speeches can be accessed through the following link and on poltextLAB’s GitHub page.

3 February, 2021 – New publication: Article by Ákos Máté, Miklós Sebők and Tamás Barczikay in PLOS ONE Ákos Máté, Miklós Sebők and Tamás Barczikay recently published an article entitled ‘The effect of central bank communication on sovereign bond yields: The case of Hungary’  The open-access article is available here: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0245515 ‘In this article, we investigate how the public communication of the Hungarian Central Bank’s Monetary Council (MC) affects Hungarian sovereign bond yields. This research ties into the advances made in the financial and political economy literature which rely on extensive textual data and quantitative text analysis tools. While prior research demonstrated that forward guidance, in the form of council meeting minutes or press releases can be used as predictors of rate decisions, we are interested in whether they are able to directly influence asset returns as well. In order to capture the effect of central bank communication, we measure the latent hawkish or dovish sentiment of MC press releases from 2005 to 2019 by applying a sentiment dictionary, a staple in the text mining toolkit. Our results show that central bank forward guidance has an intra-year effect on bond yields. However, the hawkish or dovish sentiment of press releases has no impact on maturities of one year or longer where the policy rate proves to be the most important explanatory variable. Our research also contributes to the literature by applying a specialized dictionary to monetary policy as well as broadening the discussion by analyzing a case from the non-eurozone Central-Eastern region of the European Union.’