Creating vocabulary exercises through NLP

Abstract

The use of technologies in Humanities opens new research opportunities as it allows the access to vast amounts of data such as textual corpora. As, in the Digital Humanities domain, a considerable amount of the research is done on digitised corpora, Natural Language Processing tools can be of much help in their exploitation for they help extracting linguistic information. We present a series of experiments in which we propose text transformations to generate vocabulary learning exercises based on Natural Language Processing. We describe the corpus, databases and tools we have employed in our approach and we offer an overview of a multilingual language processing pipeline. Then we present the experiments and their output. We finally discuss the strengths and shortcomings of our approach.