Maximal derivations for probabilistic strings in stochastic languages

Shmuel Peleg*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

A probabilistic string is a sequence of probability vectors. Each vector specifies a probability distribution over the possible symbols at its location in the string. In a probabilistic grammar a probability is assigned to every derivation. Given a probabilistic string and a probabilistic grammar the concept of a maximal derivation is defined. Algorithms for finding the maximal derivation for probabilistic finite state and linear grammars are given. The case where a waveform can be segmented into several possible probabilistic strings is also considered.

Original languageEnglish
Pages (from-to)290-304
Number of pages15
JournalInformation and control
Volume42
Issue number3
DOIs
StatePublished - Sep 1979
Externally publishedYes

Fingerprint

Dive into the research topics of 'Maximal derivations for probabilistic strings in stochastic languages'. Together they form a unique fingerprint.

Cite this