The years since 1998 have seen an explosion in work in biomed- ical text mining (BioNLP) of both clinical text and the biomedical literature . The work focusing on the literature has been partic- ularly stimulated by three factors. One is simply the rapid increase in the rate of publication in general, as reflected in the growth in the contents of PubMed/MEDLINE, which has been exponential. Another is the growth in the use of high-throughput assays, which commonly produce lists of genes much larger than were seen in previous experimental methods. Interpreting these gene lists typi- cally requires the digestion of large amounts of published litera- ture. Finally, the construction of model organism and other databases has been unable to keep up with the rate of discovery of the entities that they describe . Some have observed that BioNLP, serving as a curator aid, is a potential solution to this problem.
Chapman, W. W., & Cohen, K. B. (2009, October). Current issues in biomedical text mining and natural language processing. Journal of Biomedical Informatics, 42(5), 757–759. https://doi.org/10.1016/j.jbi.2009.09.001