An Intelligent Hybrid Approach for Designing Increasing Translation Invariant Morphological Operators for Time Series Forecasting

Authors:
Ricardo A. Araújo;Robson P. Sousa;Tiago A. Ferreira
Affiliations:
Center for Informatics, Federal University of Pernambuco, Brazil;Statistics and Informatics Department, Catholic University of Pernambuco, Brazil;Statistics and Informatics Department, Catholic University of Pernambuco, Brazil
Venue:
ISNN '07 Proceedings of the 4th international symposium on Neural Networks: Part II--Advances in Neural Networks
Year:
2007

Citing 5
Cited 3

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations
Time Series Analysis: Forecasting and Control

Time Series Analysis: Forecasting and Control
Neural Networks Approach to the Random Walk Dilemma of Financial Time Series

Applied Intelligence
MRL-filters: a general class of nonlinear systems and their optimal design for image processing

IEEE Transactions on Image Processing
Tuning of the structure and parameters of a neural network using an improved genetic algorithm

IEEE Transactions on Neural Networks

Swarm-based translation-invariant morphological prediction method for financial time series forecasting

Information Sciences: an International Journal
A class of hybrid morphological perceptrons with application in time series forecasting

Knowledge-Based Systems
A Morphological-Rank-Linear evolutionary method for stock market prediction

Information Sciences: an International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, an intelligent hybrid approach is presented for designing increasing translation invariant morphological operators for time series forecasting. It consists of an intelligent hybrid model composed of a Modular Morphological Neural Network (MMNN) and an improved Genetic Algorithm (GA) with optimal genetic operators to accelerate its search convergence. The improved GA searches for the minimum number of time lags for a correct time series representation, as well as by the initial weights, architecture and number of modules of the MMNN; then each element of the improved GA population is trained via Back Propagation (BP) algorithm to further improve the parameters supplied by the improved GA. An experimental analysis is conducted with the proposed method using two real world time series and five well-known performance measurements, demonstrating good performance of this kind of morphological system for time series forecasting.