| Abstract: | The problem of retrieving time series similar to a speci ed
query pattern has been recently addressed within the Case Based Rea-
soning (CBR) literature. Providing a
exible and ecient way of dealing
with such an issue would be of paramount importance in medical do-
mains, where many patient parameters are often collected in the form
of time series. In this paper, we describe a novel framework for retriev-
ing cases with time series features, relying on Temporal Abstractions.
With respect to more classical (mathematical) approaches, our frame-
work provides signi cant advantages. In particular, multi-level abstrac-
tion mechanisms and proper indexing techniques allow for
exible query
issuing, and for ecient and interactive query answering. The frame-
work is currently being applied to the hemodialysis domain. In this eld,
experimental results have shown the superiority of our approach with re-
spect to the use of a classical mathematical technique in
exibility, user
friendliness, and also quality of results.
Tests in other application domains, as well as further enhancements, are
foreseen in our future work. |
