A study of the ultrasonic vocalizations of several adult male BALB/c mice in the presence\nof a female, is undertaken in this study. A total of 179 distinct ultrasonic syllables referred to\nas â??phonemesâ? are isolated, and in the resulting dataset, k-means and agglomerative clustering\nalgorithms are implemented to group the ultrasonic vocalizations into clusters based on features\nextracted from their pitch contours. In order to find the optimal number of clusters, the elbow\nmethod was used, and nine distinct categories were obtained. Results when the k-means method was\napplied are presented through a matching matrix, while clustering results when the agglomerative\ntechnique was applied are presented as a dendrogram. The results of both methods are in line with\nthe manual annotations made by the authors, as well as with the ones presented in the literature.\nThe two methods of unsupervised analysis applied on 14 element feature vectors provide evidence\nthat vocalizations can be grouped into nine clusters, which translates into the claim that there is a\ndistinct repertoire of â??syllablesâ? or â??phonemesâ?.
Loading....