Information Technology Reference
In-Depth Information
The last Influence Diagram evaluates the maximum usefulness expected in
the application of cognitive therapies, i.e. psychological therapies based on the
fundamentals of cognitive psychology.
5 Conclusions
This research work is in its initial stage, for that reason we have focused ex-
clusively on the model and the decisions taken to design it. We believe that in
the very near future we can obtain promising results, especially when we have
a wider database of instances and more random instances in order to construct
a realer quantitative model, which will enable us to do suciently reliable and
assessable experiments. We are also convinced that Bayesian Networks can pro-
vide researchers with a powerful tool with great analytical capacity. Bayesian
Networks and in particular the model that we present in this research work can
include new variables, be they risk factors, symptoms or intermediate variables,
and it is possible to analyse mathematically the impact that these variables can
have on the diagnosis. This analytical capacity along with experts' epidemiolog-
ical studies or subjective expert assessments could enable us to characterise the
disease in its very initial stages. Furthermore, Influence Diagrams can establish
action policies in a normative way, making it possible to apply and extend the
most recent research studies in the Diagnosis of Alzheimer. Influence Diagrams
could have great potential when applying and extending the research studies
that are obtained.It should be highlighted that this research project opens a
large multidisciplinary research field.
Acknowledgements
The authors are grateful to the CiCYT for financial aid on project TIN-2010-
20845-C03-02.
References
1. Arias-Calleja, M.: Carmen: una herramienta de software librepara modelos gráficos
probabilistas (2009)
2. Bottcher, S.G., Dethlefsen, C.: Learning bayesian networks with r. In: International
Workshop on Distributed Statistical Computing, DSC 2003 (2003)
3. Cree, G.S., McRae, K.: Analyzing the factors underlying the structure and compu-
tation of the meaning of chipmunk, cherry, chisel, cheese, and cello (and many other
such concrete nouns). Journal of Experimental Psychology: General 132, 163-201
(2003)
4. Díez-Vegas, F.J.: Teoría probabilista de la decisión en medicina (2007)
5. Fernández-Galán, S., Díez-Vegas, F.J.: Modelling Dynamic Causal Interactions
with BayesianNetworks: Temporal Noisy Gates (2000)
6. Kjaerulff, U.B., Madsen, A.L.: Bayesian Networks and Influence Diagrams
7. Lacave, C.: Explicación en Redes Bayesianas (2002)
Search WWH ::
Custom Search