Open-access datasets for time
series causality discovery validation
Isabelle
Guyon, Clopinet,
California, USA
The CausalityWorkbench project provides an environment to test causal
discovery algorithms. Via a web portal (http://clopinet.com/causality),
we
provide a number of resources, including a repository of datasets,
models,
and software packages, and a virtual laboratory allowing users to
benchmark
causal discovery algorithms by performing virtual experiments to study
artificial causal systems. We regularly organize competitions. Our
repository already includes several time dependent datasets from a
variety
of domains: system biology, neurosciences, physics, manufacturing, and
marketing. We will invite new contributions and present our plan for
upcoming evaluations of causal models for time series applications.
[NIPS 2009 Causality and Time Series
Mini-Symposium]