While MDS displays dialect continua, cluster analysis can be used for classifying dialects and identifying dialect areas. Clustering is the process of dividing a set of objects into groups (clusters). In Gabmap, the objects are geographic places which are divided into groups based on their linguistic similarity.
A problem with cluster analysis is that it is a relatively unstable method. Small changes in the distance matrix can lead to large changes in the clustering results, which can make the results of clustering unreliable. Gabmap offers a number of ways to validate the results of cluster analysis. One method is to compare the results of discrete clustering to multidimensional scaling. Another method is fuzzy clustering.
General reference(s): Everitt et al. 2011, Jain, Murty, & Flynn 1999
Use in dialectology: Goebl 1982, Goebl 1984, Heeringa 2004, Mucha & Haimerl 2005, Nerbonne, Kleiweg, Heeringa, & Manni 2008, Prokić 2010, Prokić & Nerbonne 2008, Schiltz 1996
Topics: