1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/prec_rec.R
\name{recall}
\alias{recall}
\alias{recall.default}
\alias{recall.table}
\alias{precision}
\alias{precision.default}
\alias{precision.table}
\alias{precision.matrix}
\alias{F_meas}
\alias{F_meas.default}
\alias{F_meas.table}
\title{Calculate recall, precision and F values}
\usage{
recall(data, ...)
\method{recall}{table}(data, relevant = rownames(data)[1], ...)
\method{recall}{default}(data, reference, relevant = levels(reference)[1], na.rm = TRUE, ...)
precision(data, ...)
\method{precision}{default}(data, reference, relevant = levels(reference)[1], na.rm = TRUE, ...)
\method{precision}{table}(data, relevant = rownames(data)[1], ...)
F_meas(data, ...)
\method{F_meas}{default}(
data,
reference,
relevant = levels(reference)[1],
beta = 1,
na.rm = TRUE,
...
)
\method{F_meas}{table}(data, relevant = rownames(data)[1], beta = 1, ...)
}
\arguments{
\item{data}{for the default functions, a factor containing the discrete
measurements. For the \code{table} function, a table.}
\item{...}{not currently used}
\item{relevant}{a character string that defines the factor level
corresponding to the "relevant" results}
\item{reference}{a factor containing the reference values (i.e. truth)}
\item{na.rm}{a logical value indicating whether \code{NA} values should be
stripped before the computation proceeds}
\item{beta}{a numeric value used to weight precision and recall. A value of
1 is traditionally used and corresponds to the harmonic mean of the two
values but other values weight recall beta times more important than
precision.}
}
\value{
A number between 0 and 1 (or NA).
}
\description{
These functions calculate the recall, precision or F values of a measurement
system for finding/retrieving relevant documents compared to reference
results (the truth regarding relevance). The measurement and "truth" data
must have the same two possible outcomes and one of the outcomes must be
thought of as a "relevant" results.
}
\details{
The recall (aka sensitivity) is defined as the proportion of relevant
results out of the number of samples which were actually relevant. When
there are no relevant results, recall is not defined and a value of
\code{NA} is returned.
The precision is percentage of predicted truly relevant results of the total
number of predicted relevant results and characterizes the "purity in
retrieval performance" (Buckland and Gey, 1994)
The measure "F" is a combination of precision and recall (see below).
Suppose a 2x2 table with notation
\tabular{rcc}{ \tab Reference \tab \cr Predicted \tab relevant \tab
Irrelevant \cr relevant \tab A \tab B \cr Irrelevant \tab C \tab D \cr }
The formulas used here are: \deqn{recall = A/(A+C)} \deqn{precision =
A/(A+B)} \deqn{F_i = (1+i^2)*prec*recall/((i^2 * precision)+recall)}
See the references for discussions of the statistics.
}
\examples{
###################
## Data in Table 2 of Powers (2007)
lvs <- c("Relevant", "Irrelevant")
tbl_2_1_pred <- factor(rep(lvs, times = c(42, 58)), levels = lvs)
tbl_2_1_truth <- factor(c(rep(lvs, times = c(30, 12)),
rep(lvs, times = c(30, 28))),
levels = lvs)
tbl_2_1 <- table(tbl_2_1_pred, tbl_2_1_truth)
precision(tbl_2_1)
precision(data = tbl_2_1_pred, reference = tbl_2_1_truth, relevant = "Relevant")
recall(tbl_2_1)
recall(data = tbl_2_1_pred, reference = tbl_2_1_truth, relevant = "Relevant")
tbl_2_2_pred <- factor(rep(lvs, times = c(76, 24)), levels = lvs)
tbl_2_2_truth <- factor(c(rep(lvs, times = c(56, 20)),
rep(lvs, times = c(12, 12))),
levels = lvs)
tbl_2_2 <- table(tbl_2_2_pred, tbl_2_2_truth)
precision(tbl_2_2)
precision(data = tbl_2_2_pred, reference = tbl_2_2_truth, relevant = "Relevant")
recall(tbl_2_2)
recall(data = tbl_2_2_pred, reference = tbl_2_2_truth, relevant = "Relevant")
}
\references{
Kuhn, M. (2008), ``Building predictive models in R using the
caret package, '' \emph{Journal of Statistical Software},
(\doi{10.18637/jss.v028.i05}).
Buckland, M., & Gey, F. (1994). The relationship between Recall and
Precision. \emph{Journal of the American Society for Information Science},
45(1), 12-19.
Powers, D. (2007). Evaluation: From Precision, Recall and F Factor to ROC,
Informedness, Markedness and Correlation. Technical Report SIE-07-001,
Flinders University
}
\seealso{
\code{\link{confusionMatrix}}
}
\author{
Max Kuhn
}
\keyword{manip}
|
