Extracting information from archaeological texts

Abstract

To address archaeology's most pressing substantive challenges, researchers must discover, access, and extract information contained in the reports and articles that codify so much of archaeology's knowledge. These efforts will require application of existing and emerging natural language processing technologies to extensive digital corpora. Automated classification can enable development of metadata needed for the discovery of relevant documents. Although it is even more technically challenging, automated extraction of and reasoning with information from texts can provide urgently needed access to contextualized information within documents. Effective automated translation is needed for scholars to benefit from research published in other languages.