作者: Stefan I. Adalbjornsson , Johan Sward , Jonas Wallin , Andreas Jakobsson
关键词:
摘要: In this paper, we propose a method for estimating statistical periodicities in symbolic sequences. Different from other common approaches used the estimation of sequences arbitrary, finite, symbol sets, that often map sequence to numerical representation, here exploit likelihood-based formulation sparse modeling framework represent periodic behavior sequence. The resulting criterion includes restriction on cardinality solution; two approximate solutions are suggested—one greedy and one using an iterative convex relaxation strategy ease restriction. performance proposed methods illustrated both simulated real DNA data, showing notable gain as compared estimators.