A corpus-based study on images of family members

Innovative computer technology has sparked a revolutionary change in the area of languageanalysis. Individual researchers are increasingly likely to access huge corpora containingseveral hundred million words and to analyze the data using highly developed languageanalyzers via the Internet. Sketch Engine is one of such online language analyzers, whichregards the Internet as a data source for linguistic studies. The online system allows us toanalyze a wide range of Japanese texts with more than 400 million words, as well as texts inother languages. In this paper, we will examine the use of the words for family members inJapanese texts and reveal images of each member role hidden in the same.