[Taxacom] taxonomic names databases

[David Campbell]

In my experience, the "scientific names on this page" feature has high
levels of both false positives and false negatives.  However, my experience
is biased towards older works that are likely to be more challenging for
OCR, and may also have more names missed in the list used to identify
possible taxonomic usage.  The most reliable way to improve the results
would be to have a way for users to flag the problems.

Unfortunately, people do often rely on such indexing as data for
analysis.  Some molecular workers think that simply taking the oldest date
from the Paleobiology Database is good calibration for molecular clocks,
for example.  Putting everything online and linking it makes it quite easy
to do a big meta-analysis without paying any attention to data quality.  It
also facilitates better studies, of course, but checking the primary data
is a part of a reliable study.

On Mon, Aug 22, 2016 at 9:41 AM, Nadia Talent <nadia.talent at utoronto.ca>
wrote:

> Hi Stephen,
>
> Those are extreme examples, seemingly Leptus japonicus and Crepis
> japonicus were turned into the not even remotely related Cleopus japonicus.
> The second, but not the first might perhaps be an OCR phenomenon? As for
> the first, I've seen what might be a similar mechanism, but the names that
> were mixed up were plausibly synonyms, differing in the species epithet, as
> if some other database were the source of hypercorrection (Allamanda
> neriifolia was put in place of Allamanda schottii; the replacement is,
> strangely, the junior synonym).
>
> Perhaps a weevil genus name exists that an overly imaginative OCR
> corrector could mistake for Leptus, and a synonymy database lurking
> somewhere then links it to Cleopus.
>
> I would argue that synonyms do not belong in the "Scientific Names on this
> Page" list; if they are to appear at all, it should only be as a clearly
> marked option. I hope nobody is using those "Scientific Names on this Page"
> as data for analysis!
>
> Nadia
> -------------------------------
> Nadia Talent
> Research Associate,
> Royal Ontario Museum,
> Green Plant Herbarium (TRT),
> Department of Natural History,
> 100 Queen's Park,
> Toronto, Canada, M5S 2C6
> http://nadiatalent.github.io
>
> On Aug 22, 2016, at 0:52, Stephen Thorpe <stephen_thorpe at yahoo.co.nz>
> wrote:
>
> > Here is an example of what can and does go wrong with automated methods:
> http://www.biodiversitylibrary.org/name/Cleopus_japonicus_Wingelm$,_1914
> > If you click on either of the linked pages, you will find no mention of
> Cleopus japonicus, except in the "Scientific Names on this Page" field on
> the left! Something is going awry and askew!
> >
> > Stephen
> _______________________________________________
> Taxacom Mailing List
> Taxacom at mailman.nhm.ku.edu
> http://mailman.nhm.ku.edu/cgi-bin/mailman/listinfo/taxacom
> The Taxacom Archive back to 1992 may be searched at:
> http://taxacom.markmail.org
>
> Injecting Intellectual Liquidity for 29 years.
>



-- 
Dr. David Campbell
Assistant Professor, Geology
Department of Natural Sciences
Box 7270
Gardner-Webb University
Boiling Springs NC 28017