Dialectometry refers to a set of quantitative and computational methods to infer the regional distribution of dialect similarities and differences. Traditionally, dialectometric studies were based on structured inquiry data such as the maps of individual words commonly found in linguistic atlases (e.g. here). More recently, researchers have explored corpus-based approaches, where transcriptions of interviews with dialect speakers are used as raw material (e.g. here). Despite the widespread use of dialects in Norway, there are only few dialectometric studies of Norwegian, and none of them is based on atlas data. Possible goals of this thesis include:
- reviewing existing corpus-based approaches to Norwegian dialectometry, the data they are based on, and the methods they use,
- investigating the accessibility and suitability of atlas data (e.g. from Norsk målføresynopsis) for dialectometric studies,
- producing dialectometric comparisons between the two main Norwegian dialect corpora, NDC and LIA.