Clara (Chen) Shao

BigElectricity provides a new large-scale benchmark of European energy data for scalable, non-stationary multivariate time-series forecasting research. The project grew out of ML4RW (Machine Learning-Based Algorithm for Maximizing Resource Waste Reduction), which I led as principal investigator, funded by the KIT Academy for Responsible Research, Teaching, and Innovation (ARRTI), 2024 (EUR 74,000).

Alongside the benchmark, we design state-of-the-art approaches for forecasting on large-scale, non-stationary energy time series.

Published as BigElectricity: A Benchmark for Scalable Multivariate Time Series Forecasting (CIKM 2025).

This project features the excellent work of supervised students Yue Wang and Joella Wang.