A Multistrategy Approach to Classifier Learning from Time Series

Authors:
William H. Hsu;Sylvian R. Ray;David C. Wilkins
Affiliations:
National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign, Champaign, IL 61820, USA. bhsu@cis.ksu.edu;Department of Computer Science and Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA. ray@cs.uiuc.edu;Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA. dcw@uiuc.edu
Venue:
Machine Learning - Special issue on multistrategy learning
Year:
2000

Citing 28
Cited 3

Pattern recognition: human and mechanical

Pattern recognition: human and mechanical
A time-delay neural network architecture for isolated word recognition

Neural Networks
Elements of information theory

Elements of information theory
Neural networks and the bias/variance dilemma

Neural Computation
Original Contribution: Stacked generalization

Neural Networks
Original Contribution: The gamma model-A new neural model for temporal processing

Neural Networks
A theory and methodology of inductive learning

Readings in knowledge acquisition and learning
Hierarchical mixtures of experts and the EM algorithm

Neural Computation
Artificial intelligence: a modern approach

Artificial intelligence: a modern approach
The nature of statistical learning theory

The nature of statistical learning theory
Automatic synthesis of compression techniques for heterogeneous files

Software—Practice & Experience
Bagging predictors

Machine Learning
Recommender systems

Communications of the ACM
Wrappers for feature subset selection

Artificial Intelligence - Special issue on relevance
Nonlinear time series analysis

Nonlinear time series analysis
Bayesian network models for generation of crisis management training scenarios

AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Bayesian Learning for Neural Networks

Bayesian Learning for Neural Networks
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Change of Representation and Inductive Bias

Change of Representation and Inductive Bias
Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Time Series Analysis: Forecasting and Control

Time Series Analysis: Forecasting and Control
Handbook of AI

Handbook of AI
Data Mining using MLC++, A Machine Learning Library in C++

ICTAI '96 Proceedings of the 8th International Conference on Tools with Artificial Intelligence
Knowledge-guided constructive induction

Knowledge-guided constructive induction
Time series learning with probabilistic network composites

Time series learning with probabilistic network composites
A new metric-based approach to model selection

AAAI'97/IAAI'97 Proceedings of the fourteenth national conference on artificial intelligence and ninth conference on Innovative applications of artificial intelligence
Display of information for time-critical decision making

UAI'95 Proceedings of the Eleventh conference on Uncertainty in artificial intelligence

High-Performance Commercial Data Mining: A Multistrategy Machine Learning Application

Data Mining and Knowledge Discovery
Control of inductive bias in supervised learning using evolutionary computation: a wrapper-based approach

Data mining
Mining sequential causal patterns with user-specified skeletons in multi-sequence of event data

Design and application of hybrid intelligent systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present an approach to inductive concept learning usingmultiple models for time series. Our objective is to improve theefficiency and accuracy of concept learning by decomposing learningtasks that admit multiple types of learning architectures and mixtureestimation methods. The decomposition method adapts attribute subsetselection and constructive induction (cluster definition) to definenew subproblems. To these problem definitions, we can applymetric-based model selection to select from a database of learningcomponents, thereby producing a specification for supervised learningusing a mixture model. We report positive learning results usingtemporal artificial neural networks (ANNs), on a synthetic,multiattribute learning problem and on a real-world time seriesmonitoring application.