Journal of Language and Cultural Education, 2(3) ISSN 1339-4045 (print), ISSN 1339-4584 (online) SlovakEdu 104 A Comparative Study of Phonological System of Kurdish Varieties Hiwa Asadpour 1 & Maryam Mohammadi 2 Islamic Azad University, Iran 1 hiva.dandelion@gmail.com 2 nfs89k@yahoo.com Abstract In this paper a range of methods for measuring the phonetic distance between dialectal variants are described. It concerns variants of methods as wordnet method for testing lexicostatic similarities and phonostatic differences, graded map and statistical analysis of linguistic levels. In addition, all features like simple (based on atomic characters) and complex (based on feature bundles) have been studied. The dialects were compared with each other directly and indirectly via a standard dialect. The results of comparison were classified by clustering and by training of a multidimensional map. The results were compared to well established scholarship in dialectology, yielding a calibration of the methods like information visualization technique. These results indicate that computational techniques are more sensitive in feature representations of dialects and such visualizations of information have good measures of phonetic overlap of feature bundles. The results of clustering give the sharper classification, but the graded map is a nice supplement. The findings show that Kurdish has composed of different regional groups which are relate to one ancestor which it might be the proto-Kurdish language and it is not a group of languages. Key words linguistic atlas, computational linguistics language, dialect, Kurdish, Azerbaijan-e Qarbi Introduction In traditional dialectology, isoglosses are the main focus which divided maps into dialect areas. Indeed, isoglosses sometime abridge dialectal reality too much, not only when they fail to coincide, but above all, when language varieties are spread through migration or war. This paper features computational dialect comparison and classification methods. First, we will explain the area and the language which we want to deal with it through the paper then we will show comparison of variations and methods applied here. By using these visualization and computational methods, the problems signaled above are solved. On the basis of the output of the comparison methods, then we can classify the dialects. The data used for comparing dialects comes for the most part from the Linguistic Atlas of Azerbaijan-e Qarbi (LAAQ), which was compiled by Asadpour (2002-