Customizable objective function for hidden Markov Models

UBC Community, Partners, and Alumni Publications

Customizable objective function for hidden Markov Models Malhis, Nawar

Abstract

Identifying sequences with frequent patterns is a major data-mining problem in computational biology, in this work, our focus is on utilizing a HMM like model for extracting sequences with interesting frequent patterns from a set of unlabeled sequences in a heavy noise environment. First we show that the likelihood objective function for HMMs is very sensitive to noise which limits its use to labeled sequences. Then we introduce an alternative model that we call Hidden States Model, HSM, which is a HMM with customizable objective function, OF. We show empirically (on synthetic data to facilitate precise performance evaluation) how OFs can be customized to target specific information such as patterns frequency and size. Results Show HSM vastly outperformed HMM in extracting sequences with interesting patterns for both unmutated and mutated data.

Item Metadata

Title	Customizable objective function for hidden Markov Models
Creator	Malhis, Nawar
Contributor	University of British Columbia. Centre for High-Throughput Biology; Michael Smith Laboratories
Date Issued	2013
Description	Identifying sequences with frequent patterns is a major data-mining problem in computational biology, in this work, our focus is on utilizing a HMM like model for extracting sequences with interesting frequent patterns from a set of unlabeled sequences in a heavy noise environment. First we show that the likelihood objective function for HMMs is very sensitive to noise which limits its use to labeled sequences. Then we introduce an alternative model that we call Hidden States Model, HSM, which is a HMM with customizable objective function, OF. We show empirically (on synthetic data to facilitate precise performance evaluation) how OFs can be customized to target specific information such as patterns frequency and size. Results Show HSM vastly outperformed HMM in extracting sequences with interesting patterns for both unmutated and mutated data.
Subject	Hidden Markov Models; Hidden States; computational biology; data mining; discrete sequence; frequent pattern; unlabeled sequences
Genre	Article
Type	Text
Language	eng
Date Available	2013-12-03
Provider	Vancouver : University of British Columbia Library
Rights	Attribution-NonCommercial-NoDerivs 2.5 Canada
DOI	10.14288/1.0075987
URI	http://hdl.handle.net/2429/45582
Affiliation	Other UBC
Peer Review Status	Unreviewed
Scholarly Level	Researcher
Rights URI	http://creativecommons.org/licenses/by-nc-nd/2.5/ca/
Aggregated Source Repository	DSpace

Open Collections

UBC Community, Partners, and Alumni Publications