Similarities between Arabic dialects: Investigating geographical proximity

The automatic classification of Arabic dialects is an ongoing research challenge, which has been explored in recent work that defines based on increasingly limited geographic areas like cities and provinces. This paper focuses a related yet relatively unexplored topic: the effects geographical proximity located Arab countries their dialectical similarity. Our twofold, reliant on: 1) comparing textual similarities between using cosine similarity 2) measuring distance locations. We study MADAR NADI, two established datasets with from many results indicate different may fact have more than within same country, depending proximity. correlation city suggests are closer together likely to share attributes, regardless country borders. nuance provides potential for important advancements dialect because it indicates granular approach essential understanding how frame problem identification.