This page is public

This project exists to improve access to BHL scanned journals.  Our objective is to find putative taxonomic names, identified by natural language rules, and variants thrown up by OCR uncertainty, and create a look-up table so that searches of BHL can incorporate these orthographic variants automatically.

In addition we plan to work on the prolem of automatically detecting  article start and end points withn a volume, in part by looking for bibliographies.  The detection of in-text citations should help to differentiate personal names from taxon names and help us locate bibliographic lists.