Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics

Ishanu Chattopadhyay, Yicheng Wen, Asok Ray, Shashi Phoha

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

This paper presents a new pattern discovery algorithm for constructing probabilistic finite state automata (PFSA) from symbolic sequences. The new algorithm, described as Compression via Recursive Identification of Self-Similar Semantics (CRISSiS), makes use of synchronizing strings for PFSA to localize particular states and then recursively identifies the rest of the states by computing the n-step derived frequencies. We compare our algorithm to other existing algorithms, such as D-Markov and Casual-State Splitting Reconstruction (CSSR) and show both theoretically and experimentally that our algorithm captures a larger class of models.

Original languageEnglish (US)
Title of host publicationProceedings of the 2011 American Control Conference, ACC 2011
Pages125-130
Number of pages6
StatePublished - 2011
Event2011 American Control Conference, ACC 2011 - San Francisco, CA, United States
Duration: Jun 29 2011Jul 1 2011

Other

Other2011 American Control Conference, ACC 2011
CountryUnited States
CitySan Francisco, CA
Period6/29/117/1/11

Fingerprint

Semantics
Finite automata

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Chattopadhyay, I., Wen, Y., Ray, A., & Phoha, S. (2011). Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics. In Proceedings of the 2011 American Control Conference, ACC 2011 (pp. 125-130). [5991453]
Chattopadhyay, Ishanu ; Wen, Yicheng ; Ray, Asok ; Phoha, Shashi. / Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics. Proceedings of the 2011 American Control Conference, ACC 2011. 2011. pp. 125-130
@inproceedings{41a978609ae742bca0ac0c685f2fc7ea,
title = "Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics",
abstract = "This paper presents a new pattern discovery algorithm for constructing probabilistic finite state automata (PFSA) from symbolic sequences. The new algorithm, described as Compression via Recursive Identification of Self-Similar Semantics (CRISSiS), makes use of synchronizing strings for PFSA to localize particular states and then recursively identifies the rest of the states by computing the n-step derived frequencies. We compare our algorithm to other existing algorithms, such as D-Markov and Casual-State Splitting Reconstruction (CSSR) and show both theoretically and experimentally that our algorithm captures a larger class of models.",
author = "Ishanu Chattopadhyay and Yicheng Wen and Asok Ray and Shashi Phoha",
year = "2011",
language = "English (US)",
isbn = "9781457700804",
pages = "125--130",
booktitle = "Proceedings of the 2011 American Control Conference, ACC 2011",

}

Chattopadhyay, I, Wen, Y, Ray, A & Phoha, S 2011, Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics. in Proceedings of the 2011 American Control Conference, ACC 2011., 5991453, pp. 125-130, 2011 American Control Conference, ACC 2011, San Francisco, CA, United States, 6/29/11.

Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics. / Chattopadhyay, Ishanu; Wen, Yicheng; Ray, Asok; Phoha, Shashi.

Proceedings of the 2011 American Control Conference, ACC 2011. 2011. p. 125-130 5991453.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics

AU - Chattopadhyay, Ishanu

AU - Wen, Yicheng

AU - Ray, Asok

AU - Phoha, Shashi

PY - 2011

Y1 - 2011

N2 - This paper presents a new pattern discovery algorithm for constructing probabilistic finite state automata (PFSA) from symbolic sequences. The new algorithm, described as Compression via Recursive Identification of Self-Similar Semantics (CRISSiS), makes use of synchronizing strings for PFSA to localize particular states and then recursively identifies the rest of the states by computing the n-step derived frequencies. We compare our algorithm to other existing algorithms, such as D-Markov and Casual-State Splitting Reconstruction (CSSR) and show both theoretically and experimentally that our algorithm captures a larger class of models.

AB - This paper presents a new pattern discovery algorithm for constructing probabilistic finite state automata (PFSA) from symbolic sequences. The new algorithm, described as Compression via Recursive Identification of Self-Similar Semantics (CRISSiS), makes use of synchronizing strings for PFSA to localize particular states and then recursively identifies the rest of the states by computing the n-step derived frequencies. We compare our algorithm to other existing algorithms, such as D-Markov and Casual-State Splitting Reconstruction (CSSR) and show both theoretically and experimentally that our algorithm captures a larger class of models.

UR - http://www.scopus.com/inward/record.url?scp=80053151144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053151144&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781457700804

SP - 125

EP - 130

BT - Proceedings of the 2011 American Control Conference, ACC 2011

ER -

Chattopadhyay I, Wen Y, Ray A, Phoha S. Unsupervised inductive learning in symbolic sequences via recursive identification of self-similar semantics. In Proceedings of the 2011 American Control Conference, ACC 2011. 2011. p. 125-130. 5991453