Similarly, a simple logistic/softmax regression classifier can be replaced by more sophisticated classifiers [15] or alternative tree-based schemes [46] to achieve even better discrimination.
The MMES metric can be used either in this empirical statistical model or in other more sophisticated classifiers, such as logistic regression.
However, previous literature has reported that Naive Bayes frequently tends to work well, and frequently better than more sophisticated classifiers, when the data are sparse compared to the dimensionality of the problem, even when the features (e.g., in our case, the functional linkage graphs and category feature vectors) are not truly conditionally independent [34], [35].
In forthcoming projects, more sophisticated classifiers and training schemes should be exploited for the prediction.
Palindrome frequency ratios as presented here could be incorporated into more sophisticated classifiers such as self-organizing maps, Bayesian classifiers, or support vector machines.
Lee et al. [ 11] observed the same results and found that, in general, more sophisticated classifiers give better performances than do classical methods.
Exhaustive rule classifier Previous work with hot spot prediction (Tuncbag et al., 2009) has shown that simple thresholding rules, e.g. ΔSASA > 60, can be as effective as more sophisticated classifiers.
Once this high-level factors contributing to the system's phenotype are learned, more complex (e.g., non-linear or conditional) relationships between the components in the group could be learned by more sophisticated classifiers, such as BBNs or kernel SVMs (see Step 2).
Our threshold crossing of beta amplitude as the stimulation trigger affords a good starting point for algorithmic control, although this may be further improved upon in the future with more sophisticated classifiers based on the use of multiple LFP features to track clinical state.
