The paper is devoted to the construction and expansion of such linguistic resources as dictionaries and thesauri using microtask-based crowdsourcing. Approaches for microtask complexity estimation, worker ranking, and adaptive task assignment are discussed. The implementation of these approaches is available in a software system designed using the three-tier architecture.
Keywords: linguistic resource, language resource, crowdsourcing, synset, relation, microtask, computational semantics, natural language processing, human-machine system, data mining, software