Information Technology Reference
In-Depth Information
The next step is to order the nodes to compose an answer. So, with all these
nodes, the final answer would be the composition of nodes 1, 9, 4, 5, 6, and the
causal paths derived from nodes 5 and 6 as seen in the following figure:
Fig. 17.8
Automatic answer to the question
Why smoking causes death?
17.5
Conclusion
In this paper we have approached how to get answers to causal questions by mech-
anisms reflecting medical knowledge. In the pyramid of causal questions, what-q,
how-q and - at the top - why-questions have been analyzed and preprocessed using
templates. Solutions to automatically get answers for each of them are provided.
These solutions are dependent of the field selected - medical knowledge - and
even of the furnished example. Clearly other domains or examples would request
substantial improvements, although we think that the approach followed in this pa-
per is a good start. A challenge for future work will be generalize this oncoming
from a medical example to a medical domain.
Note
Points 1, 2 and 3 of this paper are largely based in the author's contribution (with J.
A. Olivas) to the ISDA 2011 Congress, November 22-24, Córdoba, Spain.
Acknowledgement.
This work was supported by the Spanish Ministry for Economy and
Innovation and by the European Regional Development Fund (ERDF/FEDER) under grant
TIN2011-29827-C02-02
References
1. Aristotle:
The metaphysics. Penguin Books, London (1998);
(translated by Hugh
Lawson-Tancred)
2. Aristotle (ed.): Aristotle: Physics. Oxford University Press, Oxford (1996); (translated
by Robin Waterfield)
