Given an underlying function f : S → {0,1} defined on the set S of gene pair instances, the learning process produces a set of learning functions Γ = { f ^ : L → 0,1 ∣ L ⊂ S } that approximate f from the train set L. The goal is to find the best approximation function from Γ having a fitness function or a classification evaluation metric.
By using the above classification metric, SARS-CoV would be considered the first member of a new group of CoV, not a member of the core group 2 cluster, although it would be placed closest to group 2 (Δm < 10.2).
DAD: Discharge abstact database; PUMF: Public use microdata file; AHRQ Agencyy for Healthcare Research and Quality; PRAM: Post randomization method; Prec: Precision; DM: Discernability metric; MD: Minimal distortion; CM: Classification metric; IG: Information gain; PL: Privacy loss.
To test the performance of the proposed metric, a classification task was performed on a database of 32 participants partitioned into three groups of approximately equal size: healthy controls, patients diagnosed with mild AD, and those with moderate-to-severe AD.
For a binary classification, this metric is defined thus: (A.1) accuracy = t p + t n t p + t n + f p + f n where t p and f p indicate true and false positives respectively, while t n and f n tally true and false negatives.
This metric is applied to the training set and requires a classification method to be used.
Therefore, the mutual information between features and classification types is a good metric to measure the utility/privacy information implied by data, since mutual information can be used to quantify the information correlation between the input and output of a classification model.
However, MVPA yields a metric (classification performance) that does not readily lend itself to quantitative comparisons across experimental conditions, brain regions or people.
In the case where a Lp distance metric classification performance is desired, the resemblance equation in (18) can easily be modified and applied to obtain an approximate performance in the Hamming domain by ζ L p = ∑ d = 1 D v i 2 d, j 2 d d - v i 1 d, j 1 d d p p ≅ ∑ d = 1 D s ^ d i 2 d - i 1 d p p ≅ ∑ d = 1 D int S d - 1 ( max ) d - int 0 ( min ) d n LSSC d + 1 H D b i 2 d d, b i 1 d d p p (19).
