Accuracy of classification algorithms based on the given list of features for the classes of points with low and high ratings are given in the Fig. 7, in a setting when there are 66% random points selected for training and the rest 34% for testing.
Since the OZ equation is a nonlinear second-kind Fredholm equation whose key feature for the class of problems of interest is the highly computationally intensive nature of the kernel, the general approach employed in this paper is thus potentially useful for similar problems with this characteristic.
The advantage of one class methods is that it eliminates guessing at the optimal features for the negative class when they are not well defined.
To determine the most distinctive features for the positive class (that is, the lignocellulose degraders), we selected all features that received a positive weight in weight vectors of the majority of the five most accurate models.
Our ITC studies on the mycobacterial CoaE reveal as yet unknown features for this class of enzymes.
The distance measure D (X) for feature X is defined as follows: (9) D (X ) = m 1 − m 2 σ 1 2 + σ 1 2 where m1 and m2 are the mean values of feature X for the class 1 and class 2, σ 1 2 and σ 2 2 are variance of feature X for the class 1 and class 2, respectively.
Thus, the mean of the feature vector for the class is μ → = [ μ 1, μ 2 ], where μ1 and μ2 are the mean values of f1(i) and f2(i), respectively.
In these plots the results for the image class "no-celiac" are shown in black, whereas the feature distances for the class "celiac" are shown in green.
They are further distinguished from single AP2 domain-containing proteins by the absence of a B3 domain, a feature reserved for the class of genes designated as RAV proteins [ 2, 5, 8].
Letters stand for the various cover classes used throughout this paper and mark some of the training locations of distinguishing geologic features for the corresponding classes.
Additionally, likelihood functions are assumed to be conditionally independent on these features for all the classes X i, thus it is p ( z x | X i ) = p ( I x | X i ) p ( R x | X i ) (16).
