Chapter 11
Applicability Domain: Towards a More
Formal Framework to Express
the Applicability of a Model
and the Confidence in Individual
Predictions
Thierry Hanser, Chris Barber, Sébastien Guesné,
Jean François Marchaland and Stéphane Werner
Abstract A common understanding of the concept of applicability domain (AD) is
that it defines the scope in which a model can make a reliable prediction; in other
words, it is the domain within which we can trust a prediction. However, in reality,
the concept of confidence in a prediction is more complex and multi-faceted; the
applicability of a model is only one aspect amongst others. In this chapter, we will
look at these different perspectives and how existing AD methods contribute to them.
We will also try to formalise a holistic approach in the context of decision-making.
Keywords Applicability domain · Decision domain · TARDIS · QSAR ·
Machine learning · Confidence modelling
Abbreviations
3D
Three Dimension
AD
Applicability Domain
DD
Decision Domain
T. Hanser (B) · C. Barber · S. Guesné · J. F. Marchaland · S. Werner
Lhasa Limited, Granary Wharf House, 2 Canal Wharf, Leeds LS11 5PS, UK
e-mail: Thierry.Hanser@lhasalimited.org
C. Barber
e-mail: Chris.Barber@lhasalimited.org
S. Guesné
e-mail: Sebastien.Guesne@lhasalimited.org
J. F. Marchaland
e-mail: Jean-Francois.Marchaland@lhasalimited.org
S. Werner
e-mail: Stephane.Werner@lhasalimited.org
© Springer Nature Switzerland AG 2019
H. Hong (ed.), Advances in Computational Toxicology, Challenges and Advances
in Computational Chemistry and Physics 30,
https://doi.org/10.1007/978-3-030-16443-0_11
215
Applicability Domain: Towards a More
Formal Framework to Express
the Applicability of a Model
and the Confidence in Individual
Predictions
Thierry Hanser, Chris Barber, Sébastien Guesné,
Jean François Marchaland and Stéphane Werner
Abstract A common understanding of the concept of applicability domain (AD) is
that it defines the scope in which a model can make a reliable prediction; in other
words, it is the domain within which we can trust a prediction. However, in reality,
the concept of confidence in a prediction is more complex and multi-faceted; the
applicability of a model is only one aspect amongst others. In this chapter, we will
look at these different perspectives and how existing AD methods contribute to them.
We will also try to formalise a holistic approach in the context of decision-making.
Keywords Applicability domain · Decision domain · TARDIS · QSAR ·
Machine learning · Confidence modelling
Abbreviations
3D
Three Dimension
AD
Applicability Domain
DD
Decision Domain
T. Hanser (B) · C. Barber · S. Guesné · J. F. Marchaland · S. Werner
Lhasa Limited, Granary Wharf House, 2 Canal Wharf, Leeds LS11 5PS, UK
e-mail: Thierry.Hanser@lhasalimited.org
C. Barber
e-mail: Chris.Barber@lhasalimited.org
S. Guesné
e-mail: Sebastien.Guesne@lhasalimited.org
J. F. Marchaland
e-mail: Jean-Francois.Marchaland@lhasalimited.org
S. Werner
e-mail: Stephane.Werner@lhasalimited.org
© Springer Nature Switzerland AG 2019
H. Hong (ed.), Advances in Computational Toxicology, Challenges and Advances
in Computational Chemistry and Physics 30,
https://doi.org/10.1007/978-3-030-16443-0_11
215