Estimating and visualizing language similarities using weighted alignment and force-directed graph layout

The paper reports several studies about quantifying language similarity via phonetic alignment of core vocabulary items (taken from Wichman’s Automated Similarity Judgement Program data base). It turns out that weighted alignment according to the Needleman-Wunsch algorithm yields best results. For visualization and data exploration purposes, we used an implementation of the Fruchterman-Reingold algorithm, a version of force directed graph layout. This software projects large amounts of data points to a twoor three-dimensional structure in such a way that groups of mutually similar items form spatial clusters. The exploratory studies conducted along these ways lead to suggestive results that provide evidence for historical relationships beyond the traditionally recognized language families.