作者: Jefrey Lijffijt , Terttu Nevalainen , Tanja Säily , Panagiotis Papapetrou , Kai Puolamäki
DOI: 10.1093/LLC/FQU064
关键词: Text corpus 、 Context (language use) 、 Wilcoxon signed-rank test 、 Pragmatics 、 British National Corpus 、 Natural language processing 、 Corpus linguistics 、 Word lists by frequency 、 Theoretical linguistics 、 Artificial intelligence 、 Computer science 、 Linguistics
摘要: … Because the bias that we observe does not solely depend on word frequency, we cannot simply use higher cut-off values in the χ 2 or log-likelihood ratio tests to correct the bias. Notably…