In Section 4, we propose supervised strategies to combine these rankings in order to obtain the best possible predictions, classification, as well as learning to rank techniques.
In order to obtain an impression of the cellular functions associated with genes in the fourth cluster, we applied two different supervised strategies.
Through a semi-supervised strategy, the new dictionary learning method, termed JSM-DKSVD, utilises all spectrums over the local neighbourhoods of labelled training pixels for discriminative dictionary learning.
Second, we developed an approach based on a semi-supervised strategy to get high confidence negative data.
In the current study, we applied a global and non-supervised strategy to explore potential key signaling events conferring RT responsiveness or resistance in NSCLC cells.
In the next section we look at a related semi-supervised strategy of predicting novel functional modules within the unclassified ORFs.
$$\end{document} We employ a supervised learning strategy where the supervision signal can come from annotated data or can be generated within the system (bootstrapping).
When the training labels are available, it is worth using a supervised PCA strategy.
In particular, candidates are verified using a supervised classification strategy over a new descriptor based on HOG features.
