I agree with the earlier comments on the scope of this question, but
would like to add one point.
The whole question of "mining" chemical patent information is, for me,
largely a non-issue when it comes to trying to derive information on new
chemical entities. It may have a place when it relates to second-use
patents or mixture/formulation/process technology, but not to my mind
when trying to analyse first disclosures of chemical compounds.
Why do I say this? Simply on the basis that so far, all these mining
tools are based around analysing the text of patent specifications. But
no matter how good the *text* mining/analysis software, it is not going
to generate meaningful results all the time that a very large percentage
of the disclosure in the patents is in the form of *non-text* material,
i.e. specific or generic structures. Suggesting that you can get this
information easily by algorithmic text-based methods is akin to the
suggestion of 20 years ago that you could do a comprehensive chemical
search using systematic nomenclature. We may have believed it once,
before sub-structure search systems arrived, but never again!
-- Stephen Adams Magister Ltd. Crown House, 231 Kings Road, Reading, RG1 4LS, GB Tel: +44 (0)118 929 9515 Fax: +44 (0)118 929 9516 e-mail: stevea@magister.co.ukRegistered in England and Wales. Company No. 3407685 Registered address : Canada House, 272 Field End Road, Eastcote, Ruislip, Middlesex HA4 9NA.
This archive was generated by hypermail 2b29 : Fri Aug 10 2001 - 15:58:03 EDT