oconnoat / DHBenelux15

#DH Benelux 2015: A method for cleaning 19th century text with examples from Transactions of the Royal Irish Academy 1800-1899

MONDAY 8 JUNE 2015 15.30 – 16.30 Parallel Paper Sessions C

Link to Abstract

Talk outline:

1. Introduction to the Journal
   1. History of the RIA
   2. Irish Context
   3. Statistics about the corpus
2. Getting from the texts to .txts
   1. JSTOR & OCR
   2. Complications: Typographical
   3. Complications: Linguistic
   4. Complications: Layout
3. Topic Modelling the Journals
   1. Cleanup Process
      1. Stop Words
      2. Spell checking
4. Results: the Topics and what we learned about the RIA
5. Results: a Process for cleaning text someone else has OCR'd
6. Conclusions