Hybrid systems of local basis functions

Authors:
Ricardo Bezerra de Andrade e Silva;Teresa Bernarda Ludermir
Affiliations:
Center for Automated Learning and Discovery, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA, USA. E-mail: rbas@cs.cmu.edu;Centro de Informática, Universidade Federal de Pernambuco, Av. Professor Luis Freire s/n, Cidade Universitária, 50740-540 Recife, PE, Brazil. E-mail: tbl@cin.ufpe.br
Venue:
Intelligent Data Analysis
Year:
2001

Citing 23
Cited 0

Probabilistic neural networks

Neural Networks
Distributed representations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Neural networks and the bias/variance dilemma

Neural Computation
C4.5: programs for machine learning

C4.5: programs for machine learning
Knowledge-based artificial neural networks

Artificial Intelligence
Fuzzy engineering

Fuzzy engineering
A Comparative Analysis of Methods for Pruning Decision Trees

IEEE Transactions on Pattern Analysis and Machine Intelligence
Representing Probabilistic Rules with Networks of GaussianBasis Functions

Machine Learning
A Review and Empirical Evaluation of Feature Weighting Methods for aClass of Lazy Learning Algorithms

Artificial Intelligence Review - Special issue on lazy learning
Data preparation for data mining

Data preparation for data mining
Separate-and-Conquer Rule Learning

Artificial Intelligence Review
A Comparison of Prediction Accuracy, Complexity, and Training Time of Thirty-Three Old and New Classification Algorithms

Machine Learning
An Algorithm for Finding Best Matches in Logarithmic Expected Time

ACM Transactions on Mathematical Software (TOMS)
Pattern Recognition and Neural Networks

Pattern Recognition and Neural Networks
Learning from Data: Concepts, Theory, and Methods

Learning from Data: Concepts, Theory, and Methods
The Role of Occam‘s Razor in Knowledge Discovery

Data Mining and Knowledge Discovery
Bump hunting in high-dimensional data

Statistics and Computing
Rule Induction with CN2: Some Recent Improvements

EWSL '91 Proceedings of the European Working Session on Machine Learning
Obtaining Simplified Rule Bases by Hybrid Learning

ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
On Objective Measures of Rule Surprisingness

PKDD '98 Proceedings of the Second European Symposium on Principles of Data Mining and Knowledge Discovery
Simplifying decision trees: A survey

The Knowledge Engineering Review
Neural Networks And Hybrid Intelligent Models: Foundations, Theory, And Applications

IEEE Transactions on Neural Networks
Decision trees can initialize radial-basis function networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Since there is no individual approach that can be universally applied to effectively solve the hard problems of artificial intelligence and data analysis, hybrid systems are necessary to better tackle specific tasks by exploiting the advantages of different methodologies in a single framework. Based on known results of combining neural networks and rule-based systems, this work presents a hybrid system with the purpose of simplifying rule sets obtained from rule induction algorithms on classification problems without increasing the accuracy error. This is motivated by assuming that simplicity can lead to more understandable models and rule induction algorithms often provide an excessive number of rules necessary to classify future examples within a given accuracy error, even after pruning. Experimental evidence suggests effective gains on a benchmark of sixteen data sets. Experiments were also performed to detect the effect of different components of the proposed approach in achieving the results and so helping to explain why this hybrid system works.