Graham Kendall
Various Images

Professor Graham Kendall

Professor Graham Kendall is the Provost and CEO of The University of Nottingham Malaysia Campus (UNMC). He is also a Pro-Vice Chancellor of the University of Nottingham.

He is a Director of MyResearch Sdn Bhd, Crops for the Future Sdn Bhd. and Nottingham Green Technologies Sdn Bhd. He is a Fellow of the British Computer Society (FBCS) and a Fellow of the Operational Research Society (FORS).

He has published over 230 peer reviewed papers. He is an Associate Editor of 10 journals and the Editor-in-Chief of the IEEE Transactions of Computational Intelligence and AI in Games.

News

I have published a number of papers on Cutting and Packing
http://bit.ly/dQPw7T
What do we spend so much in supermarkets?
http://bit.ly/1yW6If7

Latest Blog Post

Snooker: Celebrating 40 years at the Crucible

Random Blog Post

How can academics use Twitter effectively?

Publication(s)

Enumerating knight's tours using an ant colony algorithm
http://bit.ly/fMCY7C
Evolving Reusable 3D Packing Heuristics with Genetic Programming.
http://bit.ly/e75y7F
Memory Length in Hyper-heuristics: An Empirical Study
http://bit.ly/eXAo7v
Heuristic Space Diversity Management in a Meta-Hyper-Heuristic Framework
http://bit.ly/1uuQW45

Graham Kendall: Details of Requested Publication


Citation

Binner, J; Chen, Q-B. and Kendall, G Studying the Effect that a Linear Transformation has on the Time-Series Prediction Ability of an Evolutionary Neural Network. In Proceedings of the 10th Joint Conference on Information Sciences, pages 592-600, 2007.


Abstract

Artificial Neural Networks have often been used for Time Series prediction. Transforming the output data, perhaps using a linear transformation, is required when designing the network. In this paper we investigate how applying a linear transformation to the output data can affect the prediction ability of a neural network. Using six datasets from the stock market, and a population based evolutionary neural network, our experiments show that transforming the output data has a significant effect on the prediction ability of a neural network, while the in-sample prediction remains unaltered. This is an interesting finding and deserves further research to try and identify if we are able to produce better predictions of out-of-sample data by a suitable transformation of the output data.


pdf

There is no pdf available for this paper. You might like to try to obtain the original source (see the doi, for example)


doi

The doi for this publication is 10.1142/9789812709677_0086 You can link directly to the original paper, via the doi, from here

What is a doi?: A doi (Document Object Identifier) is a unique identifier for sicientific papers (and occasionally other material). This provides direct access to the location where the original article is published using the URL http://dx.doi/org/xxxx (replacing xxx with the doi). See http://dx.doi.org/ for more information



URL

This pubication does not have a URL associated with it.

The URL is only provided if there is additional information that might be useful. For example, where the entry is a book chapter, the URL might link to the book itself.


Bibtex

@INPROCEEDINGS{bck2007, author = {J. Binner and Q-B. Chen and G. Kendall},
title = {Studying the Effect that a Linear Transformation has on the Time-Series Prediction Ability of an Evolutionary Neural Network},
booktitle = {Proceedings of the 10th Joint Conference on Information Sciences},
year = {2007},
pages = {592--600},
organization = {Salt Lake City, Utah, USA, 18 - 24 July 2007},
abstract = {Artificial Neural Networks have often been used for Time Series prediction. Transforming the output data, perhaps using a linear transformation, is required when designing the network. In this paper we investigate how applying a linear transformation to the output data can affect the prediction ability of a neural network. Using six datasets from the stock market, and a population based evolutionary neural network, our experiments show that transforming the output data has a significant effect on the prediction ability of a neural network, while the in-sample prediction remains unaltered. This is an interesting finding and deserves further research to try and identify if we are able to produce better predictions of out-of-sample data by a suitable transformation of the output data.},
doi = {10.1142/9789812709677_0086},
keywords = {neural networks, artificial neural networks, time series, prediction, stock market},
owner = {gxk},
timestamp = {2011.01.01} }