@conference {776, title = {Automatic Time Series Forecasting with GRNN: A Comparison with Other Models}, booktitle = {International Work-Conference on Artificial Neural Networks}, year = {2019}, note = {TIN2015-68454-R}, month = {05/2019}, pages = {198-209}, abstract = {In this paper a methodology based on general regression neural networks for forecasting time series in an automatic way is presented. The methodology is aimed at achieving an efficient and fast tool so that a large amount of time series can be automatically predicted. In this sense, general regression neural networks present some interesting features, they have a fast single-pass learning and produce deterministic results. The methodology has been implemented in the R environment. A study of packages in R for automatic time series forecasting, including well-known statistical and computational intelligence models such as exponential smoothing, ARIMA or multilayer perceptron, is also done, together with an experimentation on running time and forecast accuracy based on data from the NN3 forecasting competition.}, keywords = {Automatic forecasting, General regression neural networks, time series forecasting}, doi = {https://doi.org/10.1007/978-3-030-20521-8_17}, author = {Francisco Mart{\'\i}nez and Francisco Charte and A.J. Rivera-Rivas and Fr{\'\i}as, Mar{\'\i}a Pilar} } @article {777, title = {predtoolsTS: R package for streamlining time series forecasting}, journal = {Progress in Artificial Intelligence}, volume = {8}, year = {2019}, note = {TIN2015-68854-R}, month = {06/2019}, pages = {505{\textendash}510}, abstract = {Time series forecasting is a field of interest in many areas. Classically, statistical methods have been used to address this problem. In recent years, machine learning (ML) algorithms have been also applied with satisfactory results. However, ML software packages are not skilled to deal with raw sequences of temporal data, and therefore, it is necessary to transform these time series. This paper presents predtoolsTS, an R package that provides a uniform interface for applying both statistical and ML methods to time series forecasting. predtoolsTS comprises four modules: preprocessing, modeling, prediction and postprocessing, in order to deal with the whole process of time series forecasting.}, keywords = {machine learning, R, time series forecasting}, doi = {https://doi.org/10.1007/s13748-019-00193-z}, author = {Francisco Charte and Alberto Vico and M.D. P{\'e}rez-Godoy and A.J. Rivera-Rivas} } @article {MARTINEZ201838, title = {Dealing with seasonality by narrowing the training set in time series forecasting with kNN}, journal = {Expert Systems with Applications}, volume = {103}, year = {2018}, pages = {38 - 48}, abstract = {In this paper, a new strategy for dealing with time series exhibiting a seasonal pattern is proposed. The strategy is applied in the context of time series forecasting using kNN regression. The key idea is to forecast every different season using a different specialized kNN learner. Each learner is specialized because its training set only contains examples whose targets belong to the season that is able to forecast. This way, the forecast of a specialized kNN learner is an aggregation of target values of the same season, reducing the likelihood of misleading forecasts. Although the strategy is applied to kNN, we think that other computational intelligence approaches could take advantage of it.}, keywords = {NN regression, Seasonal time series, time series forecasting}, issn = {0957-4174}, doi = {https://doi.org/10.1016/j.eswa.2018.03.005}, url = {http://www.sciencedirect.com/science/article/pii/S0957417418301441}, author = {Francisco Mart{\'\i}nez and Mar{\'\i}a Pilar Fr{\'\i}as and M.D. P{\'e}rez-Godoy and A.J. Rivera-Rivas} } @conference {4626756, title = {An Study on Data Mining Methods for Short-Term Forecasting of the Extra Virgin Olive Oil Price in the Spanish Market}, booktitle = {2008 Eighth International Conference on Hybrid Intelligent Systems}, year = {2008}, month = {Sep.}, pages = {943-946}, abstract = {This paper presents the adaptation of an evolutionary cooperative competitive RBFN learning algorithm, CO2RBFN, for short-term forecasting of extra virgin olive oil price. The olive oil time series has been analyzed with a new evolutionary proposal for the design of RBFNs, CO2RBFN. Results obtained has been compared with ARIMA models and other data mining methods such as a fuzzy system developed with a GA-P algorithm, a multilayer perceptron trained with a conjugate gradient algorithm and a radial basis function network trained with a LMS algorithm. The experimentation shows the high efficacy reached for the applied methods, specially for data mining methods which have slightly outperformed ARIMA methodology.}, keywords = {Algorithm design and analysis, ARIMA, ARIMA models, Artificial neural networks, autoregressive moving average processes, CO2RBFN, conjugate gradient algorithm, conjugate gradient methods, data mining, data mining methods, evolutionary cooperative competitive, extra virgin olive oil price, Forecasting, fuzzy system, Fuzzy systems, GA-P algorithm, genetic algorithms, least mean squares methods, LMS algorithm, multilayer perceptron, multilayer perceptrons, Olive Oil Price, olive oil time series, Petroleum, pricing, radial basis function network, radial basis function networks, RBFN learning algorithm, short-term forecasting, Spanish market, time series, Time series analysis, time series forecasting, Training, vegetable oils}, doi = {10.1109/HIS.2008.132}, author = {P. P{\'e}rez and M. P. Fr{\'\i}as and M.D. P{\'e}rez-Godoy and A.J. Rivera-Rivas and M. J. d. Jesus and M. Parras and F. J. Torres} }