2022/02/07 by Ilya Smirnov, Ivan P. Yamshchikov, Smirnov, Ilya +1
Computer Science · #49Q22 #Computation and Language (cs.CL) #FOS: Computer and information sciences #I.2.7 #acm:49Q22 #cs.CL #msc:49Q22
paper · pdf · doi:10.48550/arxiv.2202.03119
arxiv created 2022/02/08 · arxiv updated 2022/02/09
The word mover's distance (WMD) is a popular semantic similarity metric for two texts. This position paper studies several possible extensions of WMD. We experiment with the frequency of words in the corpus as a weighting factor and the geometry of the word vector space. We validate possible extensions of WMD on six document classification datasets. Some proposed extensions show better results in terms of the k-nearest neighbor classification error than WMD.