Semi-Markov Conditional Random Field with High-Order Features

  • NVCUONG, SG COMP.NUS.EDU., YENAN, LEEWS
  • Published 2011

Abstract

We extend first-order semi-Markov conditional random fields (semi-CRFs) to include higherorder semi-Markov features, and present efficient inference and learning algorithms, under the assumption that the higher-order semiMarkov features are sparse. We empirically demonstrate that high-order semi-CRFs outperform high-order CRFs and first-order semi-CRFs on… (More)

4 Figures and Tables

Topics

  • Presentations referencing similar topics