QML-Morven

A Novel Framework for Learning Qualitative Models

Research output: Book/ReportOther Report

5 Downloads (Pure)

Abstract

In this report, a novel qualitative model learning (QML) framework named
QML-Morven is presented. QML-Morven is an extensible framework and currently includes three QML subsystems, which employ either symbolic or evolutionary approaches
as their learning strategies. QML-Morven uses the formalism of Morven, a fuzzy qualitative simulator, to represent and reason about qualitative models, and it also utilises
Morven to verify candidate models. Based on this framework, a series of experiments were
designed and carried out to: (1) verify the results obtained by the previous QML system
ILP-QSI; (2) investigate factors that in¿uence the learning precision and minimum data
requirement for successful learning; (3) address the scalability issue of QML systems.
Original languageEnglish
Place of PublicationAberdeen
PublisherDepartment of Computing Science, University of Aberdeen
Number of pages37
Publication statusPublished - Jun 2012

Publication series

NameTechnical Report ABDN–CS–12–03

Fingerprint

Learning systems
Scalability
Simulators
Experiments

Keywords

  • qualitative reasoning
  • qualitative model learning
  • artificial immune systems
  • backtracking with forward checking

Cite this

Pang, W., & Coghill, G. M. (2012). QML-Morven: A Novel Framework for Learning Qualitative Models. (Technical Report ABDN–CS–12–03). Aberdeen: Department of Computing Science, University of Aberdeen.

QML-Morven : A Novel Framework for Learning Qualitative Models. / Pang, Wei; Coghill, George M.

Aberdeen : Department of Computing Science, University of Aberdeen, 2012. 37 p. (Technical Report ABDN–CS–12–03).

Research output: Book/ReportOther Report

Pang, W & Coghill, GM 2012, QML-Morven: A Novel Framework for Learning Qualitative Models. Technical Report ABDN–CS–12–03, Department of Computing Science, University of Aberdeen, Aberdeen.
Pang W, Coghill GM. QML-Morven: A Novel Framework for Learning Qualitative Models. Aberdeen: Department of Computing Science, University of Aberdeen, 2012. 37 p. (Technical Report ABDN–CS–12–03).
Pang, Wei ; Coghill, George M. / QML-Morven : A Novel Framework for Learning Qualitative Models. Aberdeen : Department of Computing Science, University of Aberdeen, 2012. 37 p. (Technical Report ABDN–CS–12–03).
@book{abae43296b7c416b8444198e67c86543,
title = "QML-Morven: A Novel Framework for Learning Qualitative Models",
abstract = "In this report, a novel qualitative model learning (QML) framework namedQML-Morven is presented. QML-Morven is an extensible framework and currently includes three QML subsystems, which employ either symbolic or evolutionary approachesas their learning strategies. QML-Morven uses the formalism of Morven, a fuzzy qualitative simulator, to represent and reason about qualitative models, and it also utilisesMorven to verify candidate models. Based on this framework, a series of experiments weredesigned and carried out to: (1) verify the results obtained by the previous QML systemILP-QSI; (2) investigate factors that in¿uence the learning precision and minimum datarequirement for successful learning; (3) address the scalability issue of QML systems.",
keywords = "qualitative reasoning, qualitative model learning, artificial immune systems, backtracking with forward checking",
author = "Wei Pang and Coghill, {George M.}",
year = "2012",
month = "6",
language = "English",
series = "Technical Report ABDN–CS–12–03",
publisher = "Department of Computing Science, University of Aberdeen",

}

TY - BOOK

T1 - QML-Morven

T2 - A Novel Framework for Learning Qualitative Models

AU - Pang, Wei

AU - Coghill, George M.

PY - 2012/6

Y1 - 2012/6

N2 - In this report, a novel qualitative model learning (QML) framework namedQML-Morven is presented. QML-Morven is an extensible framework and currently includes three QML subsystems, which employ either symbolic or evolutionary approachesas their learning strategies. QML-Morven uses the formalism of Morven, a fuzzy qualitative simulator, to represent and reason about qualitative models, and it also utilisesMorven to verify candidate models. Based on this framework, a series of experiments weredesigned and carried out to: (1) verify the results obtained by the previous QML systemILP-QSI; (2) investigate factors that in¿uence the learning precision and minimum datarequirement for successful learning; (3) address the scalability issue of QML systems.

AB - In this report, a novel qualitative model learning (QML) framework namedQML-Morven is presented. QML-Morven is an extensible framework and currently includes three QML subsystems, which employ either symbolic or evolutionary approachesas their learning strategies. QML-Morven uses the formalism of Morven, a fuzzy qualitative simulator, to represent and reason about qualitative models, and it also utilisesMorven to verify candidate models. Based on this framework, a series of experiments weredesigned and carried out to: (1) verify the results obtained by the previous QML systemILP-QSI; (2) investigate factors that in¿uence the learning precision and minimum datarequirement for successful learning; (3) address the scalability issue of QML systems.

KW - qualitative reasoning

KW - qualitative model learning

KW - artificial immune systems

KW - backtracking with forward checking

M3 - Other Report

T3 - Technical Report ABDN–CS–12–03

BT - QML-Morven

PB - Department of Computing Science, University of Aberdeen

CY - Aberdeen

ER -