http://weblog.ib.hu-berlin.de/?p=6861 Weblog am Institut für Bibliotheks- und Informationswissenschaft der Humboldt-Universität zu Berlin Fri, 10 May 2013 13:52:25 +0000 hourly 1 http://wordpress.org/?v=3.0.4 http://weblog.ib.hu-berlin.de/?p=6861&cpage=1#comment-703485 Martin Thu, 07 May 2009 11:21:39 +0000 http://weblog.ib.hu-berlin.de/?p=6861#comment-703485 Das sieht ja zunächst ziemlich beeindruckend aus. Aber leider: "To enable semantic mark-up of text to be applied cost-effectively in a journal publishing environment, it will be necessary to automate it. Sophisticated text mining and natural language processing tools are currently being developed to recognise textual instances and link them automatically to domain-specific ontologies. However, our own experience in marking up the text of the chosen PLoS NTD article by Reis et al. (2008) clearly showed the requirement for human intervention." (<a href="http://imageweb.zoo.ox.ac.uk/pub/2008/plospaper/latest/annotation_guidelines.html" rel="nofollow">http://imageweb.zoo.ox.ac.uk/pub/2008/plospaper/latest/annotation_guidelines.html</a>) Wenn dasselbe Ergebnis eines Tages tatsächlich automatisch erzeugt werden kann, bin ich *wirklich* beeindruckt. Das sieht ja zunächst ziemlich beeindruckend aus. Aber leider: “To enable semantic mark-up of text to be applied cost-effectively in a journal publishing environment, it will be necessary to automate it. Sophisticated text mining and natural language processing tools are currently being developed to recognise textual instances and link them automatically to domain-specific ontologies. However, our own experience in marking up the text of the chosen PLoS NTD article by Reis et al. (2008) clearly showed the requirement for human intervention.” (http://imageweb.zoo.ox.ac.uk/pub/2008/plospaper/latest/annotation_guidelines.html)
Wenn dasselbe Ergebnis eines Tages tatsächlich automatisch erzeugt werden kann, bin ich *wirklich* beeindruckt.

]]>