Remco R. Bouckaert Eibe Frank Evaluating the Replicability of Significance Tests for Comparing Learning Algorithms. 3-12 http://springerlink.metapress.com/openurl.asp?genre=article&issn=0302-9743&volume=3056&spage=3 2004 conf/pakdd/2004 PAKDD db/conf/pakdd/pakdd2004.html#BouckaertF04