SimpleTS: An Efficient and Universal Model Selection Framework for Time Series Forecasting

Yuanyuan Yao, Dimeng Li, Hailiang Jie, Lu Chen, Tianyi Li, Jie Chen, Jiaqi Wang, Feifei Li, Yunjun Gao

Publikation: Bidrag til tidsskriftKonferenceartikel i tidsskriftForskningpeer review

1 Citationer (Scopus)
18 Downloads (Pure)

Abstract

Time series forecasting, that predicts events through a sequence of time, has received increasing attention in past decades. The diverse range of time series forecasting models presents a challenge for selecting the most suitable model for a given dataset. As such, the Alibaba Cloud database monitoring system must address the issue of selecting an optimal forecasting model for a single time series data. While several model selection frameworks, including AutoAITS, have been developed to predict a dataset, their effectiveness may be limited as they may not adapt well to all types of time series, resulting in reduced prediction accuracy. Alternatively, models such as AutoForecast, which train on individual data points, may offer better adaptability but are limited by longer training time required. In this paper, we introduce SimpleTS, a versatile framework for time series forecasting that exhibits high efficiency and accuracy across all types of time series data. When performing an online prediction task, SimpleTS first classifies input time series into one type, and then efficiently selects the most suitable prediction model for this type. To optimize performance, SimpleTS (i) clusters models with similar performance to improve the efficiency of classification; (ii) uses soft labeling and weighted representation learning to achieve higher classification accuracy for different time series types. Extensive experiments on 3 private datasets and 52 public datasets show that SimpleTS outperforms the state-of-the-art toolkits in terms of both training time and prediction accuracy.

OriginalsprogEngelsk
TidsskriftProceedings of the VLDB Endowment
Vol/bind16
Udgave nummer12
Sider (fra-til)3741-3753
Antal sider13
ISSN2150-8097
DOI
StatusUdgivet - 2023
Begivenhed49th International Conference on Very Large Data Bases, VLDB 2023 - Vancouver, Canada
Varighed: 28 aug. 20231 sep. 2023

Konference

Konference49th International Conference on Very Large Data Bases, VLDB 2023
Land/OmrådeCanada
ByVancouver
Periode28/08/202301/09/2023

Bibliografisk note

Publisher Copyright:
© 2023, VLDB Endowment. All rights reserved.

Fingeraftryk

Dyk ned i forskningsemnerne om 'SimpleTS: An Efficient and Universal Model Selection Framework for Time Series Forecasting'. Sammen danner de et unikt fingeraftryk.

Citationsformater