|
NB
|
Naïve Bayes Classifier
|
|
\(RSS\)
|
Received signal strength or detection power, model parameter
|
|
\(HR\)
|
Hit ratio, model parameter
|
|
\(CRL\)
|
Consecutive record length, model parameter
|
|
\(NR\)
|
Noise ratio, model parameter
|
|
\({\delta }^{2}L\)
|
Difference in time-lab between detections, model parameter
|
|
\(PDH\)
|
Proximate detection history
|
|
\(P\left({C}_{i}|{F}_{1},\dots ,{F}_{n}\right)\)
|
Posterior probability of a detection belonging to detection class \({C}_{i}\) given the observed predictor variables \({F}_{1},\dots ,{F}_{n}.\)
|
|
\(P({C}_{i})\)
|
Prior probability of the ith detection class occurring where (\(C\in \left\{\mathrm{Valid},\mathrm{False Positive}\right\}\))
|
|
\(P({F}_{j}|{C}_{i})\)
|
Likelihood (conditional probability) of the jth observed predictor (\({F}_{j}\)) value given the ith detection class (\({C}_{i}\))
|
|
\({t}_{p}\)
|
True positive
|
|
\({f}_{p}\)
|
False positive
|
|
\({f}_{n}\)
|
False negative
|
|
\({t}_{n}\)
|
True negative
|
|
\(sen\)
|
Sensitivity, model metric, \(sen={t}_{p}/({t}_{p}+{f}_{n})\)
|
|
\(spc\)
|
Specificity, model metric,\(spc={t}_{n}/({f}_{p}+{t}_{n})\)
|
|
\(npv\)
|
Negative predictive value, model metric,\(npv={t}_{n}/({f}_{n}+{t}_{n})\)
|
|
\(ppv\)
|
Positive predictive value, model metric,\(ppv={t}_{p}/({t}_{p}+{f}_{p})\)
|
|
\(fpr\)
|
False positive rate, model metric,\(fpr={f}_{p}/({f}_{p}+{t}_{n})\)
|
|
PRC
|
Precision-recall curve
|
|
AUC
|
Area under the curve statistic, integral of the PRC
|
|
\(\kappa\)
|
Cohen’s Kappa measure of concordance
|
