Neural network ensembles for time series forecasting

Landassuri-Moreno, V.; Bullinaria, John A.

doi:10.1145/1569901.1570067

Cited by 20 publications

(18 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Please refer to work of Ruta et al [27], Zhang [36], Landassuri-Moreno et al [12], Zhang et al [35], Giles et al [7], Lai et al [11], Khashei et al [10] and Patra et al [23].…”

Section: Application Of Multilayered Perceptrons From the Aspect Of Tmentioning

confidence: 99%

A system for deductive prediction and analysis of movement of basketball referees

Pecev

Racković

Ivković

2015

Multimed Tools Appl

View full text Add to dashboard Cite

This paper presents us with an automatic prediction and analysis of basketball referees movement which is useful for educational software. Such software would be very beneficial in training the young basketball referees. The paper proposes that the movement prediction of basketball referees can be achieved with a multilayered perceptron neural network. Network will reason on the basis of a ball movement during a play action. Proposed neural network will be trained with a modified Back Propagation algorithm which essentially presents a special algorithm for a multiple dependent Time Series prediction. In this paper, we will also describe initial designs of a neural network structure that, we believe, would better suit the nature of a multiple dependent Time Series prediction problems. The aforementioned educational software is capable of determining whether a referee was moving properly in a certain situation or not. Determination is possible on the basis of numerical values that are calculated by simulating the human visual field. The referee's horizontal field of view simulation is based on the standard set by the American Optometric Association. It is implemented through a modified Sweep and Prune algorithm which is also discussed in this paper.Keywords Multiple dependent time series · MLP neural networks · LTR -MDTS model · Field of vision simulation · Basketball referee movement Predrag Pecev

show abstract

“…Please refer to work of Ruta et al [27], Zhang [36], Landassuri-Moreno et al [12], Zhang et al [35], Giles et al [7], Lai et al [11], Khashei et al [10] and Patra et al [23].…”

Section: Application Of Multilayered Perceptrons From the Aspect Of Tmentioning

confidence: 99%

A system for deductive prediction and analysis of movement of basketball referees

Pecev

Racković

Ivković

2015

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…Finally, we compare our layered bagging with standard bagging and ensemble approaches described in [65] on several traditional time series such as Henon, Ikeda, Sunspot, SP 500, and Dow Jones. We here use here the same experimental setup and performance metric as [65].…”

Section: Further Experiments and Comparisonmentioning

confidence: 99%

Layered Ensemble Architecture for Time Series Forecasting

Rahman

Islam

Murase

et al. 2016

IEEE Trans. Cybern.

View full text Add to dashboard Cite

Time series forecasting (TSF) has been widely used in many application areas such as science, engineering, and finance. The phenomena generating time series are usually unknown and information available for forecasting is only limited to the past values of the series. It is, therefore, necessary to use an appropriate number of past values, termed lag, for forecasting. This paper proposes a layered ensemble architecture (LEA) for TSF problems. Our LEA consists of two layers, each of which uses an ensemble of multilayer perceptron (MLP) networks. While the first ensemble layer tries to find an appropriate lag, the second ensemble layer employs the obtained lag for forecasting. Unlike most previous work on TSF, the proposed architecture considers both accuracy and diversity of the individual networks in constructing an ensemble. LEA trains different networks in the ensemble by using different training sets with an aim of maintaining diversity among the networks. However, it uses the appropriate lag and combines the best trained networks to construct the ensemble. This indicates LEAs emphasis on accuracy of the networks. The proposed architecture has been tested extensively on time series data of neural network (NN)3 and NN5 competitions. It has also been tested on several standard benchmark time series data. In terms of forecasting accuracy, our experimental results have revealed clearly that LEA is better than other ensemble and nonensemble methods.

show abstract

“…in the situation that each non-output node i is connected to node i + 1. For a more complete description of the algorithm, see [3], [16]- [18].…”

Section: Epnet and Modular Epnetmentioning

confidence: 99%

Biasing the evolution of modular neural networks

Landassuri-Moreno¹,

Bullinaria²

2011

2011 IEEE Congress of Evolutionary Computation (CEC)

Self Cite

View full text Add to dashboard Cite

Abstract-Modularity is known to have benefits for neural systems and their evolution, and this paper aims to improve the evolutionary neural network algorithm EPNet to take advantage of those benefits. Neural networks exist with varying degrees of modularity ranging from pure modular networks characterized by disjoint partitions of hidden nodes with no communication between modules, to pure homogeneous networks with significant connections throughout. In between are apparently homogeneous networks that can be seen to have some degree of modularity if the hidden nodes are reorganized appropriately. In this paper, a modularity measure is presented and extended that can be applied to any neuron at any level in the network to provide a fine analysis of node partitioning. It also allows the rearrangement of nodes to create modules in homogeneous networks, and that is used to improve the EPNet algorithm to evolve modular neural networks. Experimental results on a simple classification task confirm that the new modular EPNet algorithm does indeed lead to more modular networks than the classical EPNet algorithm, without compromising the performance on the given task.

show abstract

Neural network ensembles for time series forecasting

Cited by 20 publications

References 14 publications

A system for deductive prediction and analysis of movement of basketball referees

A system for deductive prediction and analysis of movement of basketball referees

Layered Ensemble Architecture for Time Series Forecasting

Biasing the evolution of modular neural networks

Contact Info

Product

Resources

About