Learning Temporal Concepts from Heterogeneous Data Sequences

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We are concerned with sequences that comprise heterogeneous symbolic data that have an underlying similar temporal pattern. The data are heterogeneous with respect to classification schemes where the class values differ between sequences. However, because the sequences relate to the same underlying concept, the mappings between values, which are not known ab initio, may be learned. Such mappings relate local ontologies, in the form of classification schemes, to a global ontology (the underlying pattern). On the basis of these mappings we use maximum likelihood techniques to learn the probabilistic description of local probabilistic concepts represented by individual temporal instances of the expression sequences. This stage is followed by one in which we learn the temporal probabilistic concept that describes the underlying pattern. Such an approach has a number of advantages: (1) it provides an intuitive way of describing the underlying temporal pattern; (2) it provides a way of mapping heterogeneous sequences; (3) it allows us to take account of natural variability in the process, via probabilistic semantics; (4) it allows us to characterise the sequences in terms of a temporal probabilistic concept model. This concept may then be matched with known genetic processes and pathways.
LanguageEnglish
Pages109-117
JournalSoft Computing
Volume8
Issue number2
DOIs
Publication statusPublished - 1 Dec 2003

Fingerprint

Ontology
Maximum likelihood
Semantics
Maximum Likelihood
Concepts
Learning
Intuitive
Pathway
Model
Class
Form

Keywords

  • Clustering
  • Sequence processing
  • Schema mapping

Cite this

@article{7e6ed312c62e40dd81044e902cfa2e47,
title = "Learning Temporal Concepts from Heterogeneous Data Sequences",
abstract = "We are concerned with sequences that comprise heterogeneous symbolic data that have an underlying similar temporal pattern. The data are heterogeneous with respect to classification schemes where the class values differ between sequences. However, because the sequences relate to the same underlying concept, the mappings between values, which are not known ab initio, may be learned. Such mappings relate local ontologies, in the form of classification schemes, to a global ontology (the underlying pattern). On the basis of these mappings we use maximum likelihood techniques to learn the probabilistic description of local probabilistic concepts represented by individual temporal instances of the expression sequences. This stage is followed by one in which we learn the temporal probabilistic concept that describes the underlying pattern. Such an approach has a number of advantages: (1) it provides an intuitive way of describing the underlying temporal pattern; (2) it provides a way of mapping heterogeneous sequences; (3) it allows us to take account of natural variability in the process, via probabilistic semantics; (4) it allows us to characterise the sequences in terms of a temporal probabilistic concept model. This concept may then be matched with known genetic processes and pathways.",
keywords = "Clustering, Sequence processing, Schema mapping",
author = "SI McClean and BW Scotney and FL Palmer",
note = "This paper develops a principled mechanism for clustering temporal patterns expressed by data from heterogeneous classification schemes, thus enabling temporal concepts to be identified even when the data are from a “mixed bag”. The approach was developed in the EU-IST MISSION project to handle data heterogeneity issues between national statistical institutes. However, it has wider applicability: in this paper, to identify significant gene expression patterns; and the work is currently being used in a project (cell-phone video streaming in care support for Alzheimer's disease), funded by the ETAC consortium, to develop algorithms to identify behavioural patterns from sensor data.",
year = "2003",
month = "12",
day = "1",
doi = "10.1007/s00500-002-0251-1",
language = "English",
volume = "8",
pages = "109--117",
journal = "Soft Computing",
issn = "1432-7643",
number = "2",

}

Learning Temporal Concepts from Heterogeneous Data Sequences. / McClean, SI; Scotney, BW; Palmer, FL.

In: Soft Computing, Vol. 8, No. 2, 01.12.2003, p. 109-117.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Learning Temporal Concepts from Heterogeneous Data Sequences

AU - McClean, SI

AU - Scotney, BW

AU - Palmer, FL

N1 - This paper develops a principled mechanism for clustering temporal patterns expressed by data from heterogeneous classification schemes, thus enabling temporal concepts to be identified even when the data are from a “mixed bag”. The approach was developed in the EU-IST MISSION project to handle data heterogeneity issues between national statistical institutes. However, it has wider applicability: in this paper, to identify significant gene expression patterns; and the work is currently being used in a project (cell-phone video streaming in care support for Alzheimer's disease), funded by the ETAC consortium, to develop algorithms to identify behavioural patterns from sensor data.

PY - 2003/12/1

Y1 - 2003/12/1

N2 - We are concerned with sequences that comprise heterogeneous symbolic data that have an underlying similar temporal pattern. The data are heterogeneous with respect to classification schemes where the class values differ between sequences. However, because the sequences relate to the same underlying concept, the mappings between values, which are not known ab initio, may be learned. Such mappings relate local ontologies, in the form of classification schemes, to a global ontology (the underlying pattern). On the basis of these mappings we use maximum likelihood techniques to learn the probabilistic description of local probabilistic concepts represented by individual temporal instances of the expression sequences. This stage is followed by one in which we learn the temporal probabilistic concept that describes the underlying pattern. Such an approach has a number of advantages: (1) it provides an intuitive way of describing the underlying temporal pattern; (2) it provides a way of mapping heterogeneous sequences; (3) it allows us to take account of natural variability in the process, via probabilistic semantics; (4) it allows us to characterise the sequences in terms of a temporal probabilistic concept model. This concept may then be matched with known genetic processes and pathways.

AB - We are concerned with sequences that comprise heterogeneous symbolic data that have an underlying similar temporal pattern. The data are heterogeneous with respect to classification schemes where the class values differ between sequences. However, because the sequences relate to the same underlying concept, the mappings between values, which are not known ab initio, may be learned. Such mappings relate local ontologies, in the form of classification schemes, to a global ontology (the underlying pattern). On the basis of these mappings we use maximum likelihood techniques to learn the probabilistic description of local probabilistic concepts represented by individual temporal instances of the expression sequences. This stage is followed by one in which we learn the temporal probabilistic concept that describes the underlying pattern. Such an approach has a number of advantages: (1) it provides an intuitive way of describing the underlying temporal pattern; (2) it provides a way of mapping heterogeneous sequences; (3) it allows us to take account of natural variability in the process, via probabilistic semantics; (4) it allows us to characterise the sequences in terms of a temporal probabilistic concept model. This concept may then be matched with known genetic processes and pathways.

KW - Clustering

KW - Sequence processing

KW - Schema mapping

U2 - 10.1007/s00500-002-0251-1

DO - 10.1007/s00500-002-0251-1

M3 - Article

VL - 8

SP - 109

EP - 117

JO - Soft Computing

T2 - Soft Computing

JF - Soft Computing

SN - 1432-7643

IS - 2

ER -