Kiralj R., Ferreira. M. M. C., "Literature and Internet Database Mining in a Study About the Word CHEMOMETRICS". Águas de Lindóia, SP, Brazil, 10-15/09/2006: 10th International Conference on Chemometrics in Analytical Chemistry (CAC-2006, CAC-X), Book of Abstracts (2006) OP27. Oral 27.

10th International Conference on Chemometrics in Analytical Chemistry OP27

Literature and Internet Database Mining in a Study About the
Word CHEMOMETRICS

Rudolf Kiralj*, Márcia M. C. Ferreira   rudolf@iqm.unicamp.br

Laboratório de Quimiometria Teórica e Aplicada, Instituto de Química, Universidade Estadual
de Campinas, Campinas – SP, 13083-970 BRAZIL

Keywords: bibliometrics and webometrics, linguistics of CHEMOMETRICS, exploratory analysis
_____________________________________________________________________________________
 

    The term chemometrics was  coined by Svante Wold first in  Swedish (kemometri)  in 1971¹ and soon in
English for a newly emerging  chemical  area  that  employed  mathematical  and  statistical  methods in the
treatment of chemical data.   Since then this term  has  entered into many languages  as  a part of  common
chemical  linguistics  and   scientific  publications,   and  gave  name   to  research   groups   and   societies,
meetings and schools,   regular faculty courses,  companies  and  software.   This work tends to investigate
some bibliometric, linguistic and sociological aspects of  the  term   chemometrics   by  means  of  database
mining and chemometric methods.
    Three series of  database minings were  performed.  In one, Web of Science  was  searched  for generic
form   “chemometr*”  in title, abstract, or keywords, and in address of publications in the period  1971-2005,
for the world  and  each  country.  In another database mining,  Google  and  Yahoo  search  engines  were
extensively employed  to  find out all online available languages  and  related countries that  used the word
chemometrics in national languages as well as in English.  The third internet surfing (Google) was directed
to  determine  relative  frequencies  of  the  previously  found forms for   chemometrics   in  English and the
national languages.
    The word  chemometrics  was found in  48 official languages, in  82 ortographic forms that ranged from
only  one  per  language  like   in  Swedish  or  Portuguese   (quimiometria)   to  maximum  six   in   English
(chemometrics, chemometry, chemiometrics, chemiometry, chemimetrics, chemimetry). English speaking
countries, especially  the  USA  and  UK,  prefer chemometrics among the forms more than other countries.
The ortographic forms with pronunciations, notably their -tri, -trics, -tria  ending varieties,  show interesting
geographical  patterns  in  Europe  that   depend  on  several   items  and   not  only  on  language  groups
(Germanic, Romance, Slavic, Baltic  etc.).
    There were found  76  countries worldwide and 36 in Europe that had participated in  3858  publications
with  word(s)   “chemometr*”   in  1975-2005.   The  number  of   publications  signed    by  officially   called
chemometric  laboratories,  groups  and  departments  from   17  countries  is  rather  modest, being   1189
in 1973-2005. It seems that many researchers do not use the word chemometrics in publications and even
not in the names of their  groups.   The  geographical distribution  of  the  first  data  set  shows  interesting
trends in the world and even more in Europe. The distribution curves for  No.  publications per country (log
form)  tend to approximate  normal distribution  curves  for the world   and  every continent in future. These
time changes are well observable in Europe,  where they are slowed down by political and socio-economic
processes in Eastern Europe.
    Datasets with five descriptors for  European countries and all countries were created:  No.   publications
with “chemometr*” (log form),  the first publication date,   No.   Google hits for   “chemometrics”   (log form),
human  development  index,   and researchers-in-research-and-development  index.   Principal Component
Analysis  resulted in  two principal components with  86%  of  the  total  variance for the two datasets.  The
countries  are  distinguished  with  respect  to  their  chemometric  activity  and  existence  of  chemometric
societies and groups.  Hierarchical Cluster Analysis  was  used  alsoin  these analyses.  The  chemometric
activity   is  not  completely  described   as   it   does   not  enumerate  chemometric  publications  withouth
“chemometr*”,   other chemometric activities,   and  all  inflected  forms  of   chemometrics   in  the national
languages.  Therefore,  the observed trends should be considered as qualitative and general, and country-
to-country comparisons should be omitted.  It can be said in general that active chemometric life  is related
to scientific/technological and general progress of a country.
    The  presented  bibliometric  and  internet-based  analyses  have  shown  interesting  past  and present
trends in chemometrics and of  chemometrics  in the world,  in the continents  and particular countries and
languages. These results may be useful for the entire chemometric community.

Acknowledgment: FAPESP
__________________________________________________________________________________________________________________________

References

¹ Kowalski B.; Brown S.; Vandeginste B. J. Chemometrics 1987, 1, 1-2.