Output details
11 - Computer Science and Informatics
University of Leeds
A First Experimental Demonstration of Massive Knowledge Infusion
<22>Unlike much NLP work that extracts facts, this work seeks rules that are implicitly encoded across corpora. It is the first conclusive demonstration of the benefits of chaining learned knowledge (and feasibility of the pre-requisite scalability). Our results were generalised and made accessible (Michael IJCAI 2009, Juba IJCAI 2013, Valiant 2013 ISBN9780465032716), rapidly becoming a stepping-stone for the implementation of actual textual entailment systems (Michael Commonsense 2013), leading to the adoption of our general approach by international research groups ([j44,c84] in http://www.informatik.uni-trier.de/~ley/pers/hd/d/Dietterich:Thomas_G=), and to invited talks and tutorials in major venues (IJCAI 2013, Chinese Computer Federation ADL, ICTAI 2012).