The limited amount of the sense annotated data is a big challenge for theword sense disambiguation task. As a solution to this problem, we propose analgorithm of automatic generation and labelling of the training collections based onthe monosemous relatives concept. In this article we explore the limits of thisalgorithm: we employ it to harvest training collections for all ambiguous nouns,verbs and adjectives presented in RuWordNet thesaurus and then evaluate the qualityof the obtained collections. We demonstrate that our approach can create high-quality labelled collections with almost full-coverage of the RuWordNet polysemouswords. Furthermore, we show that our method can be applied to the Word-in-Contexttask.
展开▼
