作者: Dawn Knight , Scott Piao , Paul Rayson , Dawn Archer , Francesca Bianchi
DOI:
关键词:
摘要: The last two decades have seen the development of various semantic lexical resources such as WordNet (Miller, 1995) and USAS lexicon (Rayson et al., 2004), which played an important role in areas natural language processing corpus-based studies. Recently, increasing efforts been devoted to extending frameworks existing knowledge cover more languages, EuroWordNet Global WordNet. In this paper, we report on construction large-scale multilingual lexicons for twelve employ unified Lancaster taxonomy provide a base automatic UCREL annotation system (USAS). Our work contributes towards goal constructing larger-scale higher-quality developing corpus tools based them. Lexical coverage is factor concerning quality performance tools, experiment focus evaluating achieved by evaluation shows that some those Finnish Italian over 90% while others need further expansion.