Abstract
The Historia Scholastica Project (HSP) since 2011 has manually transcribed and translated 70 of the 608 pages from Latin to English. Tools from the field of document analysis will accelerate the process. For instance, Character spotting [Majid, IJDAR/ICFHR 2022] is not sensitive to touching characters, does not require the presence of blank rows between lines of text, and allows for the inclusion of special symbols not common in most recognition software. Illuminated capitals and section markers will be recognized separately in a similar process. Natural Language Processing techniques will be used to expand the scribal abbreviations. The translation from Latin to English will follow, which will be used for deeper analysis of the text to explore the impact it has had on society. We hope to stimulate discussion of other applications of document analysis to other corpora.
Original language | American English |
---|---|
State | Published - 28 Sep 2023 |
Event | Studying Written Artefacts: Challenges and Perspectives Conference, Hamburg University - Hamburg, Germany Duration: 28 Sep 2023 → … |
Conference
Conference | Studying Written Artefacts: Challenges and Perspectives Conference, Hamburg University |
---|---|
Period | 28/09/23 → … |
EGS Disciplines
- European Languages and Societies
- Medieval Studies
- Electrical and Computer Engineering