Figure 2From: Automatic document classification of biological literatureOverview of the classification process. Full-text papers are taken from the Textpresso corpus and processed via SVM and phrase-base clustering. The end result is a large set of html files displaying the paper taxonomy.Back to article page