Graph Neural Networks for Model Recommendation using Time Series Data

by   Aleksandr Pletnev, et al.

Time series prediction aims to predict future values to help stakeholders make proper strategic decisions. This problem is relevant in all industries and areas, ranging from financial data to demand to forecast. However, it remains challenging for practitioners to select the appropriate model to use for forecasting tasks. With this in mind, we present a model architecture based on Graph Neural Networks to provide model recommendations for time series forecasting. We validate our approach on three relevant datasets and compare it against more than sixteen techniques. Our study shows that the proposed method performs better than target baselines and state of the art, including meta-learning. The results show the relevancy and suitability of GNN as methods for model recommendations in time series forecasting.


Discrete Graph Structure Learning for Forecasting Multiple Time Series

Time series forecasting is an extensively studied subject in statistics,...

Taylor-based Optimized Recursive Extended Exponential Smoothed Neural Networks Forecasting Method

A newly introduced method called Taylor-based Optimized Recursive Extend...

A Study of Joint Graph Inference and Forecasting

We study a recent class of models which uses graph neural networks (GNNs...

Application of Time Series Analysis to Traffic Accidents in Los Angeles

With the improvements of Los Angeles in many aspects, people in mounting...

Towards forecast techniques for business analysts of large commercial data sets using matrix factorization methods

This research article suggests that there are significant benefits in ex...

Darts: User-Friendly Modern Machine Learning for Time Series

We present Darts, a Python machine learning library for time series, wit...

Applying SVGD to Bayesian Neural Networks for Cyclical Time-Series Prediction and Inference

A regression-based BNN model is proposed to predict spatiotemporal quant...