Gazdasági elemzés strukturálatlan adatokon: természetesnyelv-feldolgozási esetek

2026. március 31.DOI: https://doi.org/10.25201/HSZ.25.1.27

Szerzői információk:

Temesvári Csanád https://orcid.org/0009-0002-5724-4287: Magyar Nemzeti Bank, elemző. E-mail:

Horváth Beáta https://orcid.org/0009-0001-0269-4881: Magyar Nemzeti Bank, vezető közgazdász elemző. E-mail:

Ónozó Lívia Réka https://orcid.org/0009-0002-7595-8531: Magyar Nemzeti Bank, felügyeleti tanácsadó; Budapesti Műszaki és Gazdaságtudományi Egyetem, PhD-hallgató. E-mail:

Absztrakt:

A gazdasági szövegek, mint az újságcikkek vagy a kiskereskedelmi terméknevek, alternatív, részletes és nagy gyakoriságú információforrást jelentenek, amelyek betekintést nyújthatnak a gazdasági trendekbe, és pontosabb, időszerűbb becsléseket tesznek lehetővé. Több mélytanulási modellt alakítottunk ki két különböző kutatási feladatra: 1) szentimentindex létrehozása, amely a pénzügyi és gazdasági cikkek három szentimentkategóriába sorolásából ered; és 2) a kiskereskedelmi terméknevek megfelelő vámtarifaszám-kategóriákba sorolása. Modelljeink egyrészt következetesen felülmúlták a kiskereskedelmi termékek klasszifikációjára szolgáló alapmodelleket, másrészt szentimentindexünk pontosan előre tudta jelezni a gazdasági visszaeséseket olyan esetekben, amikor nem álltak rendelkezésre magas gyakoriságú adatok.

Hivatkozás (APA):

Temesvári, C., Horváth, B., & Ónozó, L.R. (2026). Gazdasági elemzés strukturálatlan adatokon: természetesnyelv-feldolgozási esetek. Hitelintézeti Szemle, 25(1), 27–53. https://doi.org/10.25201/HSZ.25.1.27

PDF letöltés
The works on this site are licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Rovat:

Tanulmány

Journal of Economic Literature (JEL) kódok:

C43, C45, C60

Kulcsszavak:

természetesnyelv-feldolgozás, mélytanulás, makrogazdasági nowcasting, klasszifikáció

Felhasznált irodalom:

Aguilar, P. – Ghirelli, C. – Pacce, M. – Urtasun, A. (2021): Can news help measure economic sentiment? An application in COVID-19 times. Economics Letters, 199, 109730. https://doi.org/10.1016/j.econlet.2021.109730

Arthur, F.V. – Gyires-Tóth, B. – Debreczeni, M.I. – Ónozó, L.R. (2023): Language of the Market: NLP-Driven Sentiment Analysis of the Hungarian Economy. In: 14th IEEE International Conference on Cognitive Infocommunications (CogInfoCom), Budapest, Hungary, pp. 93–98. https://doi.org/10.1109/CogInfoCom59411.2023.10397544

Ash, E. – Hansen, S. (2023): Text Algorithms in Economics. Annual Review of Economics, 15: 659–688. https://doi.org/10.1146/annurev-economics-082222-074352

Babii, A. – Ghysels, E. – Striaukas, J. (2021): Machine Learning Time Series Regressions with an Application to Nowcasting. Journal of Business & Economic Statistics, 40(3): 1094–1106. https://doi.org/10.1080/07350015.2021.1899933

Bai, J. – Perron, P. (1998): Estimating and testing linear models with multiple structural changes. Econometrica, 66(1): 47–78. https://doi.org/10.2307/2998540

Baker, S.R. – Bloom, N. – Davis, S.J. (2016): Measuring economic policy uncertainty. Quarterly Journal of Economics, 131(4): 1593–1636. https://doi.org/10.1093/qje/qjw024

Beltagy, I. – Lo, K. – Cohan, A. (2019): SciBERT: A Pretrained Language Model for Scientific Text. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 3615–3620, Hong Kong, China. Association for Computational Linguistics. https://doi.org/10.18653/v1/D19-1371

Berndt, D.J. – Clifford, J. (1994): Using dynamic time warping to find patterns in time series. In: Proceedings of the 3rd International Conference on Knowledge Discovery and Data Mining. AAAIWS’94, AAAI Press, Seattle, WA, pp. 359–370. http://www.aaai.org/Papers/Workshops/1994/WS-94-03/WS94-03-031.pdf. Letöltés ideje: 2025. május 5.

Chen, C. – Palmer, A. – Sporleder, C. (2011): Enhancing active learning for semantic role labeling via compressed dependency trees. In: Proceedings of 5th International Joint Conference on Natural Language Processing, Chiang Mai, Thailand, pp. 183–191. https://aclanthology.org/I11-1021.pdf. Letöltés ideje: 2025. szeptember 28.

Devlin, J. – Chang, M.W. – Lee, K. – Toutanova, K. (2019): BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 1: 4171–4186. https://aclanthology.org/N19-1423.pdf. Letöltés ideje: 2024. december 10.

De Bondt, G. – Sun, Y. (2025): Enhancing GDP nowcasts with ChatGPT: a novel application of PMI news releases. Working Paper 3063, European Central Bank. https://doi.org/10.2866/2788332

Dickey, D.A. – Fuller, W.A. (1979): Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74(366a): 427–431. https://doi.org/10.1080/01621459.1979.10482531

Granger, C.W.J. (1969): Investigating causal relations by econometric models and cross-spectral methods. Econometrica, 37(3): 424–438. https://doi.org/10.2307/1912791

Gentzkow, M. – Kelly, B. – Taddy, M. (2019): Text as data. Journal of Economic Literature, 57(3): 535–574. https://doi.org/10.1257/jel.20181020

Hannan, E.J. – Quinn, B.G. (1979): The determination of the order of an autoregression. Journal of the Royal Statistical Society: Series B (Methodological), 41(2): 190–195. https://www.jstor.org/stable/2985032

Horváth Beáta – Lovics Gábor (2023): Havi munkaügyi adatok becslésének módszertana a KSH-ban. Szigma, 54(3–4): 205–226. https://doi.org/10.15170/SZIGMA.54.1190

Huang, A.H. – Wang, H. – Yang, Y. (2023): FinBERT: A Large Language Model for Extracting Information from Financial Text. Contemporary Accounting Research, 40(2): 806–841. https://doi.org/10.1111/1911-3846.12832

Jiang, S. – Pang, G. – Wu, M. – Kuang, L. (2012): An improved K-nearest-neighbor algorithm for text categorization. Expert Systems with Applications, 39(1): 1503–1509. https://doi.org/10.1016/j.eswa.2011.08.040

Kalamara, E. – Turrell, A. – Redl, C. – Kapetanios, G. – Kapadia, S. (2022): Making text count: Economic forecasting using newspaper text. Journal of Applied Econometrics, 37(5): 896–919. https://doi.org/10.1002/jae.2907

Liu, Y. – Ott, M. – Goyal, N. – Du, J. – Joshi, M. – Chen, D. et al. (2019): RoBERTa: A Robustly Optimized BERT Pretraining Approach. arXiv preprint arXiv:1907.11692. https://doi.org/10.48550/arXiv.1907.11692

Loughran, T. – McDonald, B. (2011): When Is a Liability Not a Liability? Textual Analysis, Dictionaries, and 10-Ks. The Journal of Finance, 66(1): 35–65. https://doi.org/10.1111/j.1540-6261.2010.01625.x

Nasiopoulos, D.K. – Roumeliotis, K.I. – Sakas, D.P. – Toudas, K. – Reklitis, P. (2025): Financial Sentiment Analysis and Classification: A Comparative Study of Fine-Tuned Deep Learning Models. International Journal of Financial Studies, 13(2), 75. https://doi.org/10.3390/ijfs13020075

Nemeskey Dávid Márk (2020): Natural Language Processing for Language Modeling. Ph. D. dissertation, Eötvös Loránd University, Budapest. https://doi.org/10.15476/ELTE.2020.066

Ónozó Lívia Réka – Putz Orsolya – Járási István – Gyires-Tóth Bálint (2024a): Kiskereskedelmi terméknevek kategorizálása Kombinált Nomenklatúra szerint. In: Berend, G. – Gosztolya, G. – Vincze, V. (eds.): XX. Magyar Számítógépes Nyelvészeti Konferencia. Szegedi Tudományegyetem, Szeged, Magyarország, pp. 131–144. https://m2.mtmt.hu/gui2/?mode=browse&params=publication;34560678. Letöltés ideje: 2024. december 3.

Ónozó, L.R. – Arthur, F.V. – Gyires-Tóth, B. (2024b): Leveraging LLMs for Financial News Analysis and Macroeconomic Indicator Nowcasting. In: IEEE Access, Vol. 12: 160529–160547. https://www.doi.org/10.1109/ACCESS.2024.3488363

Penedo, G. – Kydlíček, H. – Sabolčec, V. – Messmer, B – Foroutan, N. – Kargaran, A.H. et al. (2025): FineWeb2: One Pipeline to Scale Them All — Adapting Pre-Training Data Processing to Every Language. Second Conference on Language Modeling. https://openreview.net/pdf?id=jnRBe6zatP. Letöltés ideje: 2025. szeptember 13.

Proakis, J.G. – Manolakis, D.G. (2007): Digital Signal Processing: Principles, Algorithms and Applications, 3rd Edition. Prentice-Hall International, Incorporated. https://uvceee.files.wordpress.com/2016/09/digital_signal_processing_principles_algorithms_and_applications_third_edition.pdf. Letöltés ideje: 2025. augusztus 28.

Rostam, Z.R.K. – Kertész, G. (2025): Advances in Pre-trained Language Models for Domain-Specific Text Classification: A Systematic Review. ACM Transactions on Intelligent Systems and Technology, 16(6), 124. https://doi.org/10.1145/3763002

Schröder, C. – Niekler, A. – Potthast, M. (2022): Revisiting Uncertainty-based Query Strategies for Active Learning with Transformers. In: Muresan, S. – Nakov, P. – Villavicencio, A. (eds.): Findings of the Association for Computational Linguistics: ACL 2022. Association for Computational Linguistics, Dublin, Ireland, pp. 2194–2203. https://doi.org/10.18653/v1/2022.findings-acl.172

Schwarz, G. (1978): Estimating the Dimension of a Model. The Annals of Statistics, 6(2): 461–464. http://www.jstor.org/stable/2958889

Settles, B. (2011): From Theories to Queries: Active Learning in Practice. In: Guyon, I. – Cawley, G. – Dror, G. – Lemaire, V. – Statnikov, A. (eds.): Active Learning and Experimental Design workshop in conjunction with AISTATS 2010, pp. 1–18. http://proceedings.mlr.press/v16/settles11a/settles11a.pdf. Letöltés ideje: 2025. szeptember 10.

Sobrino, N.D. – Ghirelli, C. – Hurtado, S. – Pérez, J.J. – Urtasun, A. (2020): The narrative about the economy as a shadow forecast: an analysis using Banco de España quarterly reports. Working Papers 2042, Banco de España. https://www.bde.es/f/webbde/SES/Secciones/Publicaciones/PublicacionesSeriadas/DocumentosTrabajo/20/Files/dt2042e.pdf

Sparck Jones, K. (1972): A statistical Interpretation of Term Specificity and its Applications in Retrieval. Journal of Documentation, 28(1): 11–21. https://doi.org/10.1108/eb026526

Tetlock, P.C. (2007): Giving content to investor sentiment: The role of media in the stock market. Journal of Finance, 62(3): 1139–1168. https://doi.org/10.1111/j.1540-6261.2007.01232.x

Toda, H.Y. – Yamamoto, T. (1995): Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics, 66(1–2): 225–250. https://doi.org/10.1016/0304-4076(94)01616-8

Tong, H. (1978): On a threshold model. In: Chen, C. (ed.): Pattern Recognition and Signal Processing. NATO ASI Series E: Applied Sc., (29). Sijthoff & Noordhoff, Netherlands, pp. 575–586. https://www.researchgate.net/publication/246995827_On_a_Threshold_Model_in_Pattern_Recognition_and_Signal_Processing

Üveges István – Vági Renátó – Megyeri Andrea – Fülöp Anna – Nagy Dániel – Vadász J. Pál et al. (2024): Saving labeling cost by embracing Active Learning: a case study. In: Berend, G. – Gosztolya, G. – Vincze, V. (eds.): XX. Magyar Számítógépes Nyelvészeti Konferencia. Szegedi Tudományegyetem, Szeged, Magyarország, pp. 145–158. https://www.researchgate.net/publication/377730059_Saving_labeling_cost_by_embracing_Active_Learning_a_case_study

Vaswani, A. – Shazeer, N. – Parmar, N. – Uszkoreit, J. – Jones, L. – Gomez, A.N. et al. (2017): Attention is All You Need. Advances in Neural Information Processing Systems 30. Curran Associates Inc. https://proceedings.neurips.cc/paper_files/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf. Letöltés ideje: 2024. december 4.

Xia, Y. – Mukherjee, S. – Xie, Z. – Wu, J. – Li, X. – Aponte, R. et al. (2025): From Selection to Generation: A Survey of LLM-based Active Learning. In: Che, W. – Nabende, J. – Shutova, E. – Pilehvar, M.T. (eds.): Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics, Volume 1: Long Papers: 14552–14569. https://doi.org/10.18653/v1/2025.acl-long.708

Yang, Z.G. – Dodé, R. – Ferenczi, G. – Héja, E. – Jelencsik-Mátyus, K. – Kőrös, Á. et al. (2023): Jönnek a Nagyok! BERT-Large, GPT-2, GPT-3 nyelvmodellek magyar nyelvre (The Big Ones are Coming! BERT-Large, GPT-2, GPT-3 language models for Hungarian). In: 19. Magyar Számítógépes Nyelvészeti Konferencia (19th Hungarian Computational Linguistics Conference), Szegedi Tudományegyetem, Szeged, pp. 247–262. https://acta.bibl.u-szeged.hu/78417/. Letöltés ideje: 2024. december 12.

Yang, Z.G. – Laki, L.J. (2021): Improving Performance of Sentence-level Sentiment Analysis with Data Augmentation Methods. In: IEEE (ed.): 12th International Conference on Cognitive Infocommunications (CogInfoCom 2021): Proceedings. Institute of Electrical and Electronics Engineers (IEEE), pp. 417–422.

Yang, Z.G. – Váradi, T. (2023): Training Experimental Language Models with Low Resources, for the Hungarian Language. Acta Polytechnica Hungarica, 20(5): 169–188. https://doi.org/10.12700/APH.20.5.2023.5.11

Zhang, Z. – Strubell, E. – Hovy, E. (2022): A Survey of Active Learning for Natural Language Processing. In: Goldberg, Y. – Kozareva, Z. – Zhang, Y. (eds.): Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, Abu Dhabi, United Arab Emirates, pp. 6166–6190. https://doi.org/10.18653/v1/2022.emnlp-main.414