Statistical Machine Learning Models for Planetary Formation
After the pioneering development of a metrics to compare planetary systems, our research uses standard techniques like dimensionality reduction, Bayesian models, random Forests, and similar classifiers.
Dimensionality reduction
Comparing simulations of planetary system formation and real planetary systems requires a notion of similarity between two planetary systems. Discover a new similarity metric for planetary systems we developed, and its application to dimensionality reduction using T-SNE.
Random Forest
Observational survey are always limited in their telescope time. Selecting the best candidates for observations, for example to discover an Earth-like planet, requires classification schemes to rank real, partially observed, exoplanetary systems. Check-out how Random Forest Classifier can achieve very good performances in these tasks.
Ready to explore the possibilities with our statistical machine learning models?
Contact us now to learn more about how we can help you in understanding planetary formation and astronomical data