Waar het over gaat in cijfers: LSA als kwantitatieve benadering in tekst- en literatuurwetenschap

Samenvatting

The present article argues that the content analysis of literature may profit from computational techniques such as Latent Semantic Analysis (LSA). LSA is able to calculate the semantic distance between textual items by locating them in a vast multi-dimensional space. The results show remarkable similarity when compared to psychological data. LSA has not, however, been employed for content analysis in Dutch. We offer two explorative examples, one with Dutch lexical items and one with Dutch literary texts, to demonstrate that LSA also works in Dutch. At the same time, we hope to have demonstrated the usefulness of the technique in answering research questions bearing on literature.