EPX (version 1.0.4)

TOP1: Calculate TOP1

Description

The performance measure TOP1 is calculated as follows: after sorting the observations by their predicted probabilities of relevance (phat) in decreasing order so the first ranked observation has the highest probability of relevance, if the first ranked observation is truly relevant, TOP1 has a value of 1. Otherwise TOP1 is 0. If there are ties for the first rank, all the corresponding observations must be truly relevant for TOP1 to score 1.

Usage

TOP1(y, phat, ...)

Arguments

y

True (binary) response vector where 1 is the rare/relevant class.

phat

Numeric vector of estimated probabilities of relevance.

...

Further arguments passed to or from other methods.

Value

Numeric value of TOP1.

Examples

# NOT RUN {
## with ties in phat
# }
# NOT RUN {
resp <- c(0, rep(1, 99))
prob <- rep(1, 100)
TOP1(y = resp, phat = prob)  # expect 0

resp <- c(1, 1, 1, rep(0, 95), 1, 1)
prob <- c(1, 1, 1, rep(0, 97))
TOP1(y = resp, phat = prob)  # expect 1

## no ties in phat
resp <- c(0, rep(1, 99))
prob <- (100:1)*0.01
TOP1(y = resp, phat = prob)  # expect 0

resp <- c(1, rep(0, 99))
TOP1(y = resp, phat = prob)  # expect 1
# }