Prediction of Biological Time Series by Genetic Embedding
暂无分享,去创建一个
In this paper we consider a new approach for the selection of past samples to be used for prediction. Instead of using classical algorithms for estimating the embedding parameters, we will use a genetic algorithm where each individual represents a possible embedding solution. We will demonstrate that the proposed technique is particularly suited when dealing with the prediction of biological time series, aiming to improve the road safety by evidencing stress conditions or possible loss of consciousness.