-
Notifications
You must be signed in to change notification settings - Fork 4
Summer2019 Session6
Thursday May 9, 17:00 - 18:15 CEST
Convenor: Eleni Bozia (University of Florida)
YouTube link: https://www.youtube.com/watch?v=bk0flGaigr0
Slides:
This module will present text analysis and visualization tools. We will discuss the significance of such tools, focusing on three key components of language—vocabulary, grammar, and syntax, and consider the significance of such studies on various levels. Starting from simple word clouds to network analysis, stylometry, and customized metrics, the students will get to see different approaches to the text that constitute methodological advantages towards more profound understanding of language construction and authorial styles. Finally, the module will consider how such methods/methodologies can help us share opinions and scholarly work with the broader community.
A. Text and Data Visualization
https://dhs.stanford.edu/algorithmic-literacy/using-word-clouds-for-topic-modeling-results/ http://dh101.humanities.ucla.edu/?page_id=40 http://dh101.humanities.ucla.edu/?page_id=46 http://www.themacroscope.org/?page_id=362 (discussion on word clouds)
B. Stylometric Analysis
Authorship of Ronald Reagan’s Radio Addresses
http://www.stat.columbia.edu/~gelman/stuff_for_blog/Airoldi_PS_Final.pdf
Making Hit Music into Science
http://news.bbc.co.uk/2/hi/5083986.stm?ls
**C. Customized Metrics **
E. Bozia. 2016. Atticism: the language of 5th-century oratory or a quantifiable stylistic phenomenon? In Celano, G. (ed.) Special Issue on Treebanks. Open Linguistics 2.1. https://doi.org/10.1515/opli-2016-0029
Forensic Linguistics
http://uir.unisa.ac.za/bitstream/handle/10500/13324/dissertation_michell_cs.pdf?sequence=1
Deception in Instant Messaging
Stylometry with R
https://journal.r-project.org/archive/2016/RJ-2016-007/RJ-2016-007.pdf
Consider the advantages of text analysis and visualization for: 1. the teaching of languages, and 2. the possibilities of finding connections between languages.
Make your own collection of texts (different texts of the same author, or different authors, etc.) and run them through different tools, each time granulating your results.