Innovation, Serendipity 2.0, Filing Patents from Biomedical Literature Exploration 1
Biomedical information professionals practice in an environment in which bibliographic databases are numerous and generally very well-structured, by using codes (molecules, genes, and proteins) and the thesaurus. The volume of information made available through these databases is considerably vast. The most popular of these databases is Medline, which is freely available through the PubMed Website1. Medline is produced by the National Library of Medicine under the Department of Health of the United States. It has currently 17.2 million references among which more than 528,000 were added in 2008. Embase2, produced by Elsevier, offers access to over 12 million references.
The volume and rate of growth of information available can be analyzed from two perspectives. On the one hand, we can consider that information is better disseminated, so that these databases have a coverage which tends toward some exhaustiveness, especially if they are combined in the same bibliographic search. But on the other hand, the information sought may be difficult to obtain, either because of the size of such databases (needle in a haystack [GRI 02]), or because of knowledge fragmentation.
The latter concept requires some explanation. At present, our knowledge increases constantly and scientific literature increases accordingly. In order to process and upgrade the growing volume, the scientific ...