Abstract
In a previous modeling study, the automatic topic modeling method LDAvis was used to identify the thematic structure of a local discourse. Through the analysis, it became possible to determine which issues are relevant to the community in the period 2001–2020. The modeling specifications were adapted, whereby the nominalization of the medial linguistic data played a decisive role. The digital form of analysis of local discourse applied to southern Germany was applied to another local corpus and a nationwide corpus. The existing method is the basis for the configuration and validation (correlations and causalities) of a neural network with the aim of automating the process of identifying and evaluating local discourses. The results of the LDAvis modeling were validated using BERT calculations and found to be reliable. In this paper, the identified procedures are applied to new digitally accessible datasets and are tested again.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Corpus content: local media of the southern German regions Freiburg and South Baden: Badische Zeitung (www.badische-zeitung.de/), Wochenblatt (www.wzo.de/home), Baden FM Das Radio für Freiburg und Südbaden (www.baden.fm), Radio Regenbogen (www.regenbogen.de).
- 2.
Corpus content: Die Zeit (www.zeit.de), Frankfurter Allgemeine Zeitung (www.faz.net), Frankfurter Rundschau (www.fr.de), Süddeutsche Zeitung (www.sueddeutsche.de), Der Spiegel (www.spiegel.de), Bild (www.bild.de), Fokus (/www.focus.de), Die Welt (www.welt.de) und Stern (www.stern.de).
- 3.
Formally and semantically identical units are classified as doubles in the article. In case of semantic identity and formal (orthographic, grammatical) dissimilarity, variants are considered and listed separately. Cf. Coronakrise: Corona-Krise.
- 4.
Cf.: [T7A-loc (2,7%): Absage, Corona-Zahlen, Schutzmaske, Krankheit]; [T27A-loc (1,5%): Pandemie-Lage, Pandemie, Corona-Regeln, Corona-Infektionen, Hygieneregel]; [T28A-loc (1,5%): Quarantäne, Corona-Krise, Korona-Fälle, Fall, Welle]; [T41A-loc (1,2%): Krise, Lockerung, Covid-19, Fall, Beschränkung].
- 5.
Cf.: [T8A-nat (1,4%): Coronavirus, Virus, Kampf, Krankheit, Ausbreitung, Mensch, Schuld, Medikament, Covid-19, Mediziner, Verlierer, Corona, Verbreitung, Corona-Pandemie, Infektion, China, Bundesgesundheitsminister, Nachbarland, Besserung].
References
Röder, M., Both, A., Hinneburg, A.: Exploring the space of topic coherence measures. In: Proceedings of the Eighth ACM International Conference on Web Search and Data Mining, pp. 399–408 (2015)
Bubenhofer, N., Rothenhäusler, K., Affolter, K., Pajovic, D.: The linguistic construction of world: an example of visual analysis and methodological challenges. In: Scholz, R. (ed.) Quantifying Approaches to Discourse for Social Scientists. PSDS, pp. 251–284. Springer, Cham (2019). https://doi.org/10.1007/978-3-319-97370-8_9
Bubenhofer, N., Rothenhäusler, K.: “Die aussicht ist grandios!” – korpuslinguistische analyse narrativer muster in bergtourenberichten. In: Eller-Wildfeuer, N., Rössler, P., Wildfeuer, A. (eds.) Alpindeutsch. Einfluss Und Verwendung Des Deutschen Im Alpinen Raum, pp. 39–60. Edition Vulpes, Regensburg (2018)
Scharloth, J., Bubenhofer, N., Rothenhäusler, K.: Andersschreiben aus korpuslinguistischer perspektive: datengeleitete zugänge zum stil. In: Schuster, B.-M., Tophinke, D. (eds.) Andersschreiben: Formen, Funktionen, Traditionen, pp. 157–78. Erich Schmidt Verlag, Berlin (2012)
Devlin, J., Chang, M.-W., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding (2019)
Vasawi, A., et al.: Attention is all you need (2017)
Angelov, D.: Top2Vec: distributed representations of topics (2020)
Campello, R.J.G.B., Moulavi, D., Sander, J.: Density-based clustering based on hierarchical density estimates. In: Pei, J., Tseng, V.S., Cao, L., Motoda, H., Xu, G. (eds.) PAKDD 2013. LNCS (LNAI), vol. 7819, pp. 160–172. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-37456-2_14
Qaiser, S., Ali, R.: Text mining: use of TF-IDF to examine the relevance of words to documents. Int. J. Comput. Appl. 181, 25–29 (2018)
https://experience.arcgis.com/experience/478220a4c454480e823b17327b2bf1d4/page/page_1/
Acknowledgment
This research work is kindly supported by the Dr. K.H. Eberle Foundation, Germany.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Walterscheid, H., Matoshi, V., Wiśniewiecka-Brückner, K., Rothenhäusler, K., Eckardt, F. (2021). LDAvis, BERT: Comparison of Method Application Under Corona Conditions. In: Ahram, T.Z., Karwowski, W., Kalra, J. (eds) Advances in Artificial Intelligence, Software and Systems Engineering. AHFE 2021. Lecture Notes in Networks and Systems, vol 271. Springer, Cham. https://doi.org/10.1007/978-3-030-80624-8_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-80624-8_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-80623-1
Online ISBN: 978-3-030-80624-8
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)