What is it about?
The graph-based technique finds groups among dialects based on (pronunciation) similarity, just as hundreds of other algorithms do, but it also identifies the features it uses to distinguish the groups.
Featured Image
Why is it important?
This illustrates the clustering technique well and it introduces a measure to ascertain how well a given feature is characteristic of a set of dialects, attending both to its generality within the group and its distinctiveness with respect dialects not in the group.
Read the Original
This page is a summary of: Bipartite spectral graph partitioning for clustering dialect varieties and detecting their linguistic features, Computer Speech & Language, July 2011, Elsevier,
DOI: 10.1016/j.csl.2010.05.004.
You can read the full text:
Contributors
The following have contributed to this page
