Fully automatic acquisition of taxonomic knowledge from large corpora of texts: Limited syntax knowledge representation system based on natural language

Abstract

The paper presents a new method for fully automatic knowledge acquisition from large corpora of unseen texts. The approach exploits simple, efficiently, fast and reliably extractable, parsable and indepth interpretable constructs of natural language specialized to convey taxonomic knowledge. It allows one to acquire large quantities of high quality general-purpose knowledge and practically eliminates costly and error-prone human pre&post-processing. Examples of the systemacquired concepts are discussed.KeywordsLarge CorpusWall Street JournalScrub TyphusSpotted FeverTime MagazineThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.