Learning Qualitative Models of Dynamic Systems

Authors:
David T. Hau;Enrico W. Coiera
Affiliations:
School of Medicine, The Johns Hopkins University, Baltimore, MD 21205, USA/ E-mail: dave@welchlink.welch.jhu.edu;Hewlett-Packard Laboratories, Filton Road, Stoke Gifford, Bristol BS12 6QZ, UK/ E-mail: ewc@hplb.hpl.hp.com
Venue:
Machine Learning - special issue on inductive logic programming
Year:
1997

Citing 14
Cited 18

A theory of the learnable

Communications of the ACM
Qualitative simulation

Artificial Intelligence
Scaling Theorems for Zero Crossings

IEEE Transactions on Pattern Analysis and Machine Intelligence
Uniqueness of the Gaussian Kernel for Scale-Space Filtering

IEEE Transactions on Pattern Analysis and Machine Intelligence
Circuits,signals,and systems

Circuits,signals,and systems
Occam's razor

Information Processing Letters
Wavelets and dilation equations: a brief introduction

SIAM Review
Discrete-time signal processing

Discrete-time signal processing
Continuing adventures in qualitative modeling—a qualitative heart model

IEA/AIE '90 Proceedings of the 3rd international conference on Industrial and engineering applications of artificial intelligence and expert systems - Volume 1
An introduction to computational learning theory

An introduction to computational learning theory
Qualitative system identification: deriving structure from behavior

Artificial Intelligence
Learning Logical Definitions from Relations

Machine Learning
Learning Decision Lists

Machine Learning
Integrating Quantitative and Qualitative Discovery: The ABACUS System

Machine Learning

Challenges for Inductive Logic Programming

EPIA '99 Proceedings of the 9th Portuguese Conference on Artificial Intelligence: Progress in Artificial Intelligence
Learning in Clausal Logic: A Perspective on Inductive Logic Programming

Computational Logic: Logic Programming and Beyond, Essays in Honour of Robert A. Kowalski, Part I
Machine Learning in Medical Applications

Machine Learning and Its Applications, Advanced Lectures
Learning qualitative models

AI Magazine
Qualitatively faithful quantitative prediction

Artificial Intelligence
Qualitative simulation and related approaches for the analysis of dynamic systems

The Knowledge Engineering Review
Modified clonal selection algorithm for learning qualitative compartmental models of metabolic systems

Proceedings of the 9th annual conference companion on Genetic and evolutionary computation
Data mining with Temporal Abstractions: learning rules from time series

Data Mining and Knowledge Discovery
Learning Qualitative Models of Physical and Biological Systems

Computational Discovery of Scientific Knowledge
Incremental Identification of Qualitative Models of Biological Systems using Inductive Logic Programming

The Journal of Machine Learning Research
Qualitative system identification from imperfect data

Journal of Artificial Intelligence Research
Temporal data mining for the quality assessment of hemodialysis services

Artificial Intelligence in Medicine
Using temporal constraints for temporal abstraction

Journal of Intelligent Information Systems
Review:

The Knowledge Engineering Review
An immune-inspired approach to qualitative system identification of biological pathways

Natural Computing: an international journal
Learning qualitative models from numerical data

Artificial Intelligence
A new approach to the abstraction of monitoring data in intensive care

AIME'05 Proceedings of the 10th conference on Artificial Intelligence in Medicine
Learning qualitative models from numerical data: extended abstract

IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

The automated construction of dynamic system modelsis an important application area for ILP. We describe amethod that learns qualitative models from time-varying physiologicalsignals. The goal is to understand the complexity of thelearning task when faced with numerical data, what signal processing techniques are required, and how this affects learning. The qualitative representation is based on Kuipers‘ QSIM. Thelearning algorithm for model construction is based on Coiera‘s GENMODEL. We show that QSIM models are efficiently PAClearnable from positive examples only, and that GENMODEL is anILP algorithm for efficiently constructing a QSIM model. We describe bothGENMOEL which performs RLGG on qualitative states to learn aQSIM model, and the front-end processing and segmenting stagesthat transform a signal into a set of qualitative states.Next we describe results of experiments on data from six cardiac bypass patients. Useful models wereobtained, representing both normal and abnormal physiologicalstates. Model variation across time and across different levels oftemporal abstraction and fault tolerance is explored.The assumption made by many previous workers that the abstraction of examples from data can be separated from the learning task is not supported by this study. Firstly, the effects ofnoise in the numerical data manifest themselvesin the qualitative examples. Secondly, the models learned aredirectly dependent on the initial qualitative abstraction chosen.