An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors
In this paper a novel approach for automatically configuring a k-nearest neighbors regressor for univariate time series forecasting is presented. The approach uses an ensemble consisting of several k-nearest neighbors models with different configurations for their hyperparameters and model selection...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10820325/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841550820452073472 |
|---|---|
| author | Maria P. Frias Francisco Martinez |
| author_facet | Maria P. Frias Francisco Martinez |
| author_sort | Maria P. Frias |
| collection | DOAJ |
| description | In this paper a novel approach for automatically configuring a k-nearest neighbors regressor for univariate time series forecasting is presented. The approach uses an ensemble consisting of several k-nearest neighbors models with different configurations for their hyperparameters and model selection choices. One advantage of this scheme is that the uncertainty associated with choosing a wrong configuration for the model is reduced. This approach is compared with the classical way of selecting a configuration by doing a grid search among several configurations of hyperparameters and model selection choices and choosing the one that performs best on a validation set. The experimental results, using datasets from time series forecasting competitions, show that, in line with previous works, the use of an ensemble produces a robust model, outperforming the approach that uses a grid search for obtaining the best configuration on a validation set and almost any specific configuration. The forecast accuracy of the ensemble is similar to state-of-the-art models. Furthermore, this paper also tests the effectiveness of some recent approaches for dealing with trending time series when using the k-nearest neighbors algorithm. |
| format | Article |
| id | doaj-art-10159edab0034e6490dd2686d8033cf8 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-10159edab0034e6490dd2686d8033cf82025-01-10T00:00:54ZengIEEEIEEE Access2169-35362025-01-01134117412510.1109/ACCESS.2025.352556110820325An Ensemble for Automatic Time Series Forecasting With K-Nearest NeighborsMaria P. Frias0https://orcid.org/0000-0001-6886-0953Francisco Martinez1https://orcid.org/0000-0002-5206-1898Department of Statistics and Operations Research, University of Jaén, Jaén, SpainDepartment of Computer Science, Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Jaén, Jaén, SpainIn this paper a novel approach for automatically configuring a k-nearest neighbors regressor for univariate time series forecasting is presented. The approach uses an ensemble consisting of several k-nearest neighbors models with different configurations for their hyperparameters and model selection choices. One advantage of this scheme is that the uncertainty associated with choosing a wrong configuration for the model is reduced. This approach is compared with the classical way of selecting a configuration by doing a grid search among several configurations of hyperparameters and model selection choices and choosing the one that performs best on a validation set. The experimental results, using datasets from time series forecasting competitions, show that, in line with previous works, the use of an ensemble produces a robust model, outperforming the approach that uses a grid search for obtaining the best configuration on a validation set and almost any specific configuration. The forecast accuracy of the ensemble is similar to state-of-the-art models. Furthermore, this paper also tests the effectiveness of some recent approaches for dealing with trending time series when using the k-nearest neighbors algorithm.https://ieeexplore.ieee.org/document/10820325/Nearest neighborstrending time seriesunivariate time series forecasting |
| spellingShingle | Maria P. Frias Francisco Martinez An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors IEEE Access Nearest neighbors trending time series univariate time series forecasting |
| title | An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors |
| title_full | An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors |
| title_fullStr | An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors |
| title_full_unstemmed | An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors |
| title_short | An Ensemble for Automatic Time Series Forecasting With K-Nearest Neighbors |
| title_sort | ensemble for automatic time series forecasting with k nearest neighbors |
| topic | Nearest neighbors trending time series univariate time series forecasting |
| url | https://ieeexplore.ieee.org/document/10820325/ |
| work_keys_str_mv | AT mariapfrias anensembleforautomatictimeseriesforecastingwithknearestneighbors AT franciscomartinez anensembleforautomatictimeseriesforecastingwithknearestneighbors AT mariapfrias ensembleforautomatictimeseriesforecastingwithknearestneighbors AT franciscomartinez ensembleforautomatictimeseriesforecastingwithknearestneighbors |