Planning spelling normalization: typology of graphic variation in the Parish Memoirs (1758) corpus

Digital Humanities are now essential for studies on large-scale textual corpora, where the transformation of text into processable data regarding linguistic phenomena requires a multidisciplinary treatment. In this article we will present an approach in Digital Humanities, which was applied to a Portuguese textual corpus from the 18th-century, gathered from a set of documents known as Memorias Paroquiais ["The Parish Memoirs"], with high historical and heritage value. We will highlight some corpus constitution characteristics, questions concerning the expressive spelling variation perceived in the texts. We propose a typology towards a future automatic normalization of this textual corpus.