Although a comprehensive comparison is difficult, a comparison of high level characteristics is possible.
Consequently, when working together with low-level characteristics, high-level characteristics can improve the classification results by considering more than the physical attributes.
Besides, another important characteristic is that all of these low- and high-level characteristics are linked into the same meta-model, allowing traceability among them.
Our analysis of large mutation datasets shows that B-SIFT is easily scalable in the way that SIFT is, and the distribution of B-SIFT scores can be used to discover high-level characteristics of the dataset.
Instead, higher-level characteristics are extracted and used to recognize maneuvers, which makes it easier to generalize the learnt model to arbitrary layouts.
Such properties often express higher-level characteristics of the behaviour of the process rather than the direct influences; for example stable end situations (e.g., 'after some time the crime rate stabilises'), or the effect of the occurrence of several events at the same time.
We started with lower level characteristics (resident) before entering higher-level characteristics (interviewer) and the interviewer × resident interaction.
As an alternative approach, we focus on higher-level characteristics such as the known functional associations and the homo-AA tract composition of TNR sequences.
To make the use of pharmacogenomic biomarkers more clinically effective, the potentially large and complex data yielded by genotyping or sequencing need to be reduced to more manageable, higher-level characteristics such as alleles, haplotypes or phenotypes that can help to predict drug response.
These results suggest that among white women, if there is a relation of PCB exposure with breast cancer risk, it is likely quite modest, even at the relatively high levels characteristic of occupationally exposed populations.
