The World of Woman in the Bank of English: Internal Criteria for the Classification of Corpora
Junsaku Nakamura · Literary and Linguistic Computing · 1995
The extended version of Hayashi's Quantification Method Type III, one of the statistical methods for multivariate analysis, has been used by Nakamura in various studies in the field of corpus linguistics to determine the structure of large corpora and proved to be quite effective. The same method can also be applied to the study of collocations and the present article demonstrates this. The frequencies of statistically significant collocates of the word ’woman‘ in four components of the Bank of English were counted and processed by Hyashi's Method to determine the structure of the corpus as a whole based upon the distribution of collocates of ‘Woman’ in the four big components. The result shows how the word ‘woman’ is collocationally treated in the Bank of English by making explicit the relationships among the four components, the relationships among collocates and the relationships between components and collocates. The results give promise that this may be a powerful tool in the description and classification of corpora.