StreamEnsemble: Predictive Queries over Spatiotemporal Streaming Data

TL;DR

StreamEnsemble is a novel method for predictive queries over spatiotemporal streaming data that dynamically selects and allocates machine learning models based on data distribution, significantly improving accuracy and efficiency.

Contribution

It introduces a dynamic ensemble approach tailored for spatiotemporal data, addressing the limitations of single models and traditional ensembles.

Findings

Outperforms traditional ensemble methods and single models in accuracy.

Reduces prediction error by more than 10 times.

Demonstrates significant improvements in prediction time.

Abstract

Predictive queries over spatiotemporal (ST) stream data pose significant data processing and analysis challenges. ST data streams involve a set of time series whose data distributions may vary in space and time, exhibiting multiple distinct patterns. In this context, assuming a single machine learning model would adequately handle such variations is likely to lead to failure. To address this challenge, we propose StreamEnsemble, a novel approach to predictive queries over ST data that dynamically selects and allocates Machine Learning models according to the underlying time series distributions and model characteristics. Our experimental evaluation reveals that this method markedly outperforms traditional ensemble methods and single model approaches in terms of accuracy and time, demonstrating a significant reduction in prediction error of more than 10 times compared to traditional approaches.