Some Alternatives to Parikh Matrices Using String Kernels

摘要: We describe methods of representing strings as real valued vectors or matrices; we show how to integrate two separate lines enquiry: string kernels, developed in machine learning, and Parikh matrices [8], which have been studied intensively over the last few years a powerful tool study combinatorics words. In field there is widespread use analogous mappings into high dimensional feature spaces based on occurrences subwords factors. this paper one can kernels construct alternatives matrices, that overcome some limitations matrix construction. These are morphisms from free monoid rings real-valued under multiplication: subsequence kernel other gap-weighted kernel. For latter demonstrate for many values gap-weight hyperparameter resulting morphism injective.