Structured and Unstructured: Extracting Information from Classics Scholarly Texts
The project is an ongoing Computational Linguistic and Text Analytic study of how the language and structure of explicitly encoded data sources can be used to help mining texts of unencoded corpora.
The two corpora being currently considered contain respectively OCRed journal papers and working papers about Classical(Greek and Latin) texts.
The presented project aims at showing how - and with which gain in terms of accuracy - information extracted from structured data sources can be used to automatically extract information from an unstructured corpus. The extracted information is meant to be used in order to provide semantic access over the corpus itself.