Extending Corpus-Based Discourse Analysis for Exploring Japanese Social Media
Heinrich P, Schäfer F (2018)
Publication Language: English
Publication Type: Conference contribution, Conference Contribution
Publication year: 2018
Pages Range: 135 - 140
Conference Proceedings Title: Proceedings of 4th Asia Pacific Corpus Linguistics Conference (APCLC2018)
Event location: Takamatsu
Abstract
The present paper presents and applies novel techniques for corpus-based discourse analysis in order to analyze Japanese social media data. Firstly, we visualize collocational profiles of topic nodes (e. g. lexical items or corpus queries) by projecting word embeddings onto a two-dimensional semantically structured plane, where the size of the displayed items represent their association strength to the topic node. Secondly, we generalize collocations in two ways: by looking at collocates of discourses (which are created by combining several sets of lexical items), and by looking at second-order topic collocates of lexical items. We perform a discourse analysis in a corpus of more than 400,000,000 tweets produced before and in the aftermath of the Fukushima Daiichi nuclear disaster with a focus on discussions about (phasing out) nuclear energy.
Authors with CRIS profile
Philipp Heinrich
Lehrstuhl für Korpus- und Computerlinguistik
Fabian Schäfer
Lehrstuhl für Japanologie mit dem Schwerpunkt Japan der Moderne und Gegenwart
Related research project(s)
How to cite
APA:
Heinrich, P., & Schäfer, F. (2018). Extending Corpus-Based Discourse Analysis for Exploring Japanese Social Media. In Yukio Tono & Hitoshi Isahara (Eds.), Proceedings of 4th Asia Pacific Corpus Linguistics Conference (APCLC2018) (pp. 135 - 140). Takamatsu, JP.
MLA:
Heinrich, Philipp, and Fabian Schäfer. "Extending Corpus-Based Discourse Analysis for Exploring Japanese Social Media." Proceedings of the 4th Asia Pacific Corpus Linguistics Conference (APCLC2018), Takamatsu Ed. Yukio Tono & Hitoshi Isahara, 2018. 135 - 140.
BibTeX: Download