Towards the Automated Selection of ML Models for Time-Series Data Forecasting

Yi Chen, Verena Kantere

2025

Abstract

Analyzing and forecasting time-series data is challenging since the latter always comes with characteristics, such as seasonality, which may impact models’ performance but are frequently unknown before implementing models. At the same time, the abundance of ML models makes it difficult to select a suitable model for a specific dataset. To solve this problem, research is currently exploring the creation of automated model selection techniques. However, the characteristics of the datasets have yet to be considered. Toward this goal, this work aims to explore the appropriateness of models concerning the features of time-series datasets. We collect a wide range of models and time-series datasets and choose some of them to conduct experiments to explore how different elements affect the performances of selected models. Based on the results, we formulate several outcomes that are helpful in time-series data forecasting. Further, we design a decision tree based on these outcomes, which can be used as a first step toward creating an automated model-selection technique for time-series forecasting.

Paper Citation

in Harvard Style

Chen Y. and Kantere V. (2025). Towards the Automated Selection of ML Models for Time-Series Data Forecasting. In Proceedings of the 27th International Conference on Enterprise Information Systems - Volume 1: ICEIS; ISBN 978-989-758-749-8, SciTePress, pages 813-819. DOI: 10.5220/0013296100003929

in Bibtex Style

@conference{iceis25,
author={Yi Chen and Verena Kantere},
title={Towards the Automated Selection of ML Models for Time-Series Data Forecasting},
booktitle={Proceedings of the 27th International Conference on Enterprise Information Systems - Volume 1: ICEIS},
year={2025},
pages={813-819},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013296100003929},
isbn={978-989-758-749-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 27th International Conference on Enterprise Information Systems - Volume 1: ICEIS
TI - Towards the Automated Selection of ML Models for Time-Series Data Forecasting
SN - 978-989-758-749-8
AU - Chen Y.
AU - Kantere V.
PY - 2025
SP - 813
EP - 819
DO - 10.5220/0013296100003929
PB - SciTePress