| Abstract: | The problem of retrieving time series similar to a specified
query pattern has been recently addressed within the Case Based Reasoning
(CBR) literature. Providing a flexible and efficient way of dealing
with such an issue is of paramount importance in medical domains, where
many patient parameters are often collected in the form of time series. In
the past, we have developed a framework for retrieving cases with time
series features, relying on Temporal Abstractions. With respect to more
classical (mathematical) approaches, our framework provides significant
advantages. In particular, multi-level abstraction mechanisms and proper
indexing techniques allow for flexible query issuing, and for efficient and
interactive query answering. In this paper, we present an extension to
such a framework, aimed at supporting sub-series matching as well. Indeed,
sub-series retrieval may be crucial in medical applications, when
the whole time series evolution is sometimes not of interest, while critical
patterns to be search for are only “local”. Moreover, their relative order,
but not their precise location in time, may be known, and an interactive
search, at different abstraction levels, may be of great help for the
medical decision maker. The framework is currently being applied to the
hemodialysis domain. |
