Code examples for the paper:
Bing Liu, Wynne Hsu and Yiming Ma, Integrating Classification and Association Rule Mining, ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 1998.
set.seed(1234)
library("tidyverse")## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──## ✓ ggplot2 3.3.5 ✓ purrr 0.3.4
## ✓ tibble 3.1.6 ✓ dplyr 1.0.7
## ✓ tidyr 1.1.4 ✓ stringr 1.4.0
## ✓ readr 2.1.0 ✓ forcats 0.5.1## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library("arulesCBA")## Loading required package: Matrix##
## Attaching package: 'Matrix'## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpack## Loading required package: arules##
## Attaching package: 'arules'## The following object is masked from 'package:dplyr':
##
## recode## The following objects are masked from 'package:base':
##
## abbreviate, write##
## Attaching package: 'arulesCBA'## The following object is masked from 'package:arules':
##
## ruleslibrary("arulesViz")
library("caret") # for confusionMatrix## Loading required package: lattice##
## Attaching package: 'caret'## The following object is masked from 'package:purrr':
##
## liftdata(iris)Train/test split. Train classifier on 90% of the data
train <- sample(1:nrow(iris), size = as.integer(nrow(iris) * .9))Discretize using Fayyad and Irani’s MDL method
iris.disc <- iris %>% discretizeDF.supervised(Species ~ ., data = ., method = "mdlp")
iris.disc %>% head()## Sepal.Length Sepal.Width Petal.Length Petal.Width Species
## 1 [-Inf,5.55) [3.35, Inf] [-Inf,2.45) [-Inf,0.8) setosa
## 2 [-Inf,5.55) [2.95,3.35) [-Inf,2.45) [-Inf,0.8) setosa
## 3 [-Inf,5.55) [2.95,3.35) [-Inf,2.45) [-Inf,0.8) setosa
## 4 [-Inf,5.55) [2.95,3.35) [-Inf,2.45) [-Inf,0.8) setosa
## 5 [-Inf,5.55) [3.35, Inf] [-Inf,2.45) [-Inf,0.8) setosa
## 6 [-Inf,5.55) [3.35, Inf] [-Inf,2.45) [-Inf,0.8) setosairis.transactions <- transactions(iris.disc)What is the support that requires a minimum of 5 observations?
5 / length(train)## [1] 0.03703704Let’s just
cars <- mineCARs(Species ~ ., transactions = iris.transactions[train,] , support = 0.04)## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.5 0.1 1 none FALSE FALSE 5 0.04 1
## maxlen target ext
## 5 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 5
##
## set item appearances ...[15 item(s)] done [0.00s].
## set transactions ...[15 item(s), 135 transaction(s)] done [0.00s].
## sorting and recoding items ... [15 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 5 done [0.00s].
## writing ... [82 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].cars <- cars[!is.redundant(cars)]
rulebase <- cars %>% sort(by = "confidence")
inspectDT(rulebase)classifier <- CBA_ruleset(Species ~ ., rulebase)
classifier## CBA Classifier Object
## Formula: Species ~ .
## Number of rules: 26
## Default Class: NA
## Classification method: first
## Description: Custom rule setpr <- predict(classifier, iris.transactions[-train,])
confusionMatrix(reference = iris[-train, ]$Species, data = pr)## Confusion Matrix and Statistics
##
## Reference
## Prediction setosa versicolor virginica
## setosa 7 0 0
## versicolor 0 3 1
## virginica 0 2 2
##
## Overall Statistics
##
## Accuracy : 0.8
## 95% CI : (0.5191, 0.9567)
## No Information Rate : 0.4667
## P-Value [Acc > NIR] : 0.009047
##
## Kappa : 0.6875
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: setosa Class: versicolor Class: virginica
## Sensitivity 1.0000 0.6000 0.6667
## Specificity 1.0000 0.9000 0.8333
## Pos Pred Value 1.0000 0.7500 0.5000
## Neg Pred Value 1.0000 0.8182 0.9091
## Prevalence 0.4667 0.3333 0.2000
## Detection Rate 0.4667 0.2000 0.1333
## Detection Prevalence 0.4667 0.2667 0.2667
## Balanced Accuracy 1.0000 0.7500 0.7500This is not a very good classifier and it uses many rules. CBA reduces the number of rules in the rule base to make a better and more compact classifier using the missclassification error.
This prepares the data using Fayyad and Irani’s MDL discretization method
cba_model <-
CBA(Species ~ .,
data = iris[train, ],
supp = 0.03,
conf = 0.5,
disc.method = "mdlp",
verbose = TRUE)##
## Mining CARs...
## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.5 0.1 1 none FALSE FALSE 5 0.03 1
## maxlen target ext
## 5 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 4
##
## set item appearances ...[15 item(s)] done [0.00s].
## set transactions ...[15 item(s), 135 transaction(s)] done [0.00s].
## sorting and recoding items ... [15 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 5 done [0.00s].
## writing ... [98 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
##
## Pruning CARs...
## CARs left: 7cba_model## CBA Classifier Object
## Formula: Species ~ .
## Number of rules: 7
## Default Class: NA
## Classification method: first
## Description: CBA algorithm (Liu et al., 1998)CBA chooses only these rules
inspectDT(rules(cba_model))Make predictions for the test data
pr <- predict(cba_model, newdata = iris[-train, ])
pr## [1] setosa setosa setosa setosa setosa setosa
## [7] setosa virginica versicolor versicolor virginica versicolor
## [13] versicolor virginica virginica
## Levels: setosa versicolor virginicaconfusionMatrix(reference = iris[-train, ]$Species, data = pr)## Confusion Matrix and Statistics
##
## Reference
## Prediction setosa versicolor virginica
## setosa 7 0 0
## versicolor 0 3 1
## virginica 0 2 2
##
## Overall Statistics
##
## Accuracy : 0.8
## 95% CI : (0.5191, 0.9567)
## No Information Rate : 0.4667
## P-Value [Acc > NIR] : 0.009047
##
## Kappa : 0.6875
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: setosa Class: versicolor Class: virginica
## Sensitivity 1.0000 0.6000 0.6667
## Specificity 1.0000 0.9000 0.8333
## Pos Pred Value 1.0000 0.7500 0.5000
## Neg Pred Value 1.0000 0.8182 0.9091
## Prevalence 0.4667 0.3333 0.2000
## Detection Rate 0.4667 0.2000 0.1333
## Detection Prevalence 0.4667 0.2667 0.2667
## Balanced Accuracy 1.0000 0.7500 0.7500data(Zoo, package = "mlbench")
head(Zoo)## hair feathers eggs milk airborne aquatic predator toothed backbone
## aardvark TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE
## antelope TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE
## bass FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE TRUE
## bear TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE
## boar TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE
## buffalo TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE
## breathes venomous fins legs tail domestic catsize type
## aardvark TRUE FALSE FALSE 4 FALSE FALSE TRUE mammal
## antelope TRUE FALSE FALSE 4 TRUE FALSE TRUE mammal
## bass FALSE FALSE TRUE 0 TRUE FALSE FALSE fish
## bear TRUE FALSE FALSE 4 FALSE FALSE TRUE mammal
## boar TRUE FALSE FALSE 4 TRUE FALSE TRUE mammal
## buffalo TRUE FALSE FALSE 4 TRUE FALSE TRUE mammaldiscretize legs manually
Zoo$legs <- Zoo$legs > 0Train classifier on 90% of the data
train <- sample(1:nrow(Zoo), size = as.integer(nrow(Zoo) * .9))Note: CBA could use a lower minimum support.
models <- c(CBA, C4.5_CBA, FOIL, RIPPER_CBA, PART_CBA)
accuracy <- list()
nrules <- list()
for (m in models) {
cat("\n---------------------------------------------------------------------\n\n")
fit <- m(type ~ ., data = Zoo[train, ])
print(fit)
cat("\nRule base:\n")
inspect(rules(fit))
pr <- predict(fit, newdata = Zoo[-train, ])
cat("\n")
cm <- confusionMatrix(reference = Zoo[-train, ]$type, data = pr)
print(cm)
accuracy[[fit$description]] <- cm$overall["Accuracy"]
nrules[[fit$description]] <- length(rules(fit))
length(rules(fit))
}##
## ---------------------------------------------------------------------
##
## CBA Classifier Object
## Formula: type ~ .
## Number of rules: 4
## Default Class: NA
## Classification method: first
## Description: CBA algorithm (Liu et al., 1998)
##
##
## Rule base:
## lhs rhs support confidence
## [1] {milk=TRUE} => {type=mammal} 0.4000000 1.0
## [2] {feathers=TRUE} => {type=bird} 0.2222222 1.0
## [3] {breathes=FALSE, fins=TRUE} => {type=fish} 0.1111111 1.0
## [4] {} => {type=mollusc.et.al} 0.1000000 0.1
## coverage lift count size coveredTransactions totalErrors
## [1] 0.4000000 2.5 36 2 36 34
## [2] 0.2222222 4.5 20 2 20 24
## [3] 0.1111111 9.0 10 3 10 15
## [4] 1.0000000 1.0 90 1 24 15
##
## Confusion Matrix and Statistics
##
## Reference
## Prediction mammal bird reptile fish amphibian insect mollusc.et.al
## mammal 5 0 0 0 0 0 0
## bird 0 0 0 0 0 0 0
## reptile 0 0 0 0 0 0 0
## fish 0 0 0 3 0 0 0
## amphibian 0 0 0 0 0 0 0
## insect 0 0 0 0 0 0 0
## mollusc.et.al 0 0 0 0 1 1 1
##
## Overall Statistics
##
## Accuracy : 0.8182
## 95% CI : (0.4822, 0.9772)
## No Information Rate : 0.4545
## P-Value [Acc > NIR] : 0.01598
##
## Kappa : 0.7381
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: mammal Class: bird Class: reptile Class: fish
## Sensitivity 1.0000 NA NA 1.0000
## Specificity 1.0000 1 1 1.0000
## Pos Pred Value 1.0000 NA NA 1.0000
## Neg Pred Value 1.0000 NA NA 1.0000
## Prevalence 0.4545 0 0 0.2727
## Detection Rate 0.4545 0 0 0.2727
## Detection Prevalence 0.4545 0 0 0.2727
## Balanced Accuracy 1.0000 NA NA 1.0000
## Class: amphibian Class: insect Class: mollusc.et.al
## Sensitivity 0.00000 0.00000 1.00000
## Specificity 1.00000 1.00000 0.80000
## Pos Pred Value NaN NaN 0.33333
## Neg Pred Value 0.90909 0.90909 1.00000
## Prevalence 0.09091 0.09091 0.09091
## Detection Rate 0.00000 0.00000 0.09091
## Detection Prevalence 0.00000 0.00000 0.27273
## Balanced Accuracy 0.50000 0.50000 0.90000
##
## ---------------------------------------------------------------------
##
## CBA Classifier Object
## Formula: type ~ .
## Number of rules: 9
## Default Class: mollusc.et.al
## Classification method: first
## Description: RWeka classifier weka/classifiers/trees/J48
##
##
## Rule base:
## lhs rhs support confidence
## [1] {feathers=TRUE} => {type=bird} 0.22222222 1.0000000
## [2] {feathers=FALSE,
## milk=TRUE} => {type=mammal} 0.40000000 1.0000000
## [3] {feathers=FALSE,
## milk=FALSE,
## backbone=TRUE,
## fins=TRUE} => {type=fish} 0.11111111 1.0000000
## [4] {feathers=FALSE,
## milk=FALSE,
## backbone=TRUE,
## fins=FALSE,
## tail=TRUE} => {type=reptile} 0.05555556 0.8333333
## [5] {feathers=FALSE,
## milk=FALSE,
## backbone=TRUE,
## fins=FALSE,
## tail=FALSE} => {type=amphibian} 0.02222222 1.0000000
## [6] {feathers=FALSE,
## milk=FALSE,
## airborne=TRUE,
## backbone=TRUE,
## backbone=FALSE,
## tail=FALSE} => {type=insect} 0.00000000 NaN
## [7] {feathers=FALSE,
## milk=FALSE,
## airborne=FALSE,
## predator=TRUE,
## backbone=TRUE,
## backbone=FALSE,
## fins=FALSE} => {type=mollusc.et.al} 0.00000000 NaN
## [8] {feathers=FALSE,
## milk=FALSE,
## airborne=FALSE,
## predator=FALSE,
## backbone=TRUE,
## backbone=FALSE,
## fins=FALSE,
## legs=TRUE} => {type=insect} 0.00000000 NaN
## [9] {feathers=FALSE,
## milk=FALSE,
## airborne=FALSE,
## predator=FALSE,
## backbone=TRUE,
## backbone=FALSE,
## fins=FALSE,
## legs=FALSE} => {type=mollusc.et.al} 0.00000000 NaN
##
## Confusion Matrix and Statistics
##
## Reference
## Prediction mammal bird reptile fish amphibian insect mollusc.et.al
## mammal 5 0 0 0 0 0 0
## bird 0 0 0 0 0 0 0
## reptile 0 0 0 0 0 0 0
## fish 0 0 0 3 0 0 0
## amphibian 0 0 0 0 1 0 0
## insect 0 0 0 0 0 0 0
## mollusc.et.al 0 0 0 0 0 1 1
##
## Overall Statistics
##
## Accuracy : 0.9091
## 95% CI : (0.5872, 0.9977)
## No Information Rate : 0.4545
## P-Value [Acc > NIR] : 0.00243
##
## Kappa : 0.869
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: mammal Class: bird Class: reptile Class: fish
## Sensitivity 1.0000 NA NA 1.0000
## Specificity 1.0000 1 1 1.0000
## Pos Pred Value 1.0000 NA NA 1.0000
## Neg Pred Value 1.0000 NA NA 1.0000
## Prevalence 0.4545 0 0 0.2727
## Detection Rate 0.4545 0 0 0.2727
## Detection Prevalence 0.4545 0 0 0.2727
## Balanced Accuracy 1.0000 NA NA 1.0000
## Class: amphibian Class: insect Class: mollusc.et.al
## Sensitivity 1.00000 0.00000 1.00000
## Specificity 1.00000 1.00000 0.90000
## Pos Pred Value 1.00000 NaN 0.50000
## Neg Pred Value 1.00000 0.90909 1.00000
## Prevalence 0.09091 0.09091 0.09091
## Detection Rate 0.09091 0.00000 0.09091
## Detection Prevalence 0.09091 0.00000 0.18182
## Balanced Accuracy 1.00000 0.50000 0.95000
##
## ---------------------------------------------------------------------
##
## CBA Classifier Object
## Formula: type ~ .
## Number of rules: 13
## Default Class: mammal
## Classification method: weighted - using best 5 rules
## Description: FOIL-based classifier (Yin and Han, 2003)
##
##
## Rule base:
## lhs rhs support confidence lift laplace
## [1] {milk=TRUE} => {type=mammal} 0.40000000 1.0000000 2.50000 0.8604651
## [2] {feathers=TRUE} => {type=bird} 0.22222222 1.0000000 4.50000 0.7777778
## [3] {eggs=TRUE,
## fins=TRUE} => {type=fish} 0.11111111 1.0000000 9.00000 0.6470588
## [4] {predator=FALSE,
## backbone=FALSE,
## legs=TRUE} => {type=insect} 0.06666667 1.0000000 12.85714 0.5384615
## [5] {aquatic=TRUE,
## backbone=FALSE} => {type=mollusc.et.al} 0.06666667 1.0000000 10.00000 0.5384615
## [6] {airborne=TRUE,
## backbone=FALSE} => {type=insect} 0.05555556 1.0000000 12.85714 0.5000000
## [7] {hair=FALSE,
## aquatic=FALSE,
## toothed=TRUE} => {type=reptile} 0.03333333 1.0000000 18.00000 0.4000000
## [8] {backbone=FALSE,
## legs=FALSE} => {type=mollusc.et.al} 0.03333333 1.0000000 10.00000 0.4000000
## [9] {toothed=TRUE,
## venomous=TRUE,
## fins=FALSE} => {type=reptile} 0.02222222 1.0000000 18.00000 0.3333333
## [10] {aquatic=TRUE,
## breathes=TRUE,
## tail=FALSE} => {type=amphibian} 0.02222222 0.6666667 20.00000 0.3000000
## [11] {aquatic=TRUE,
## breathes=TRUE,
## catsize=FALSE} => {type=amphibian} 0.03333333 0.4285714 12.85714 0.2857143
## [12] {feathers=FALSE,
## toothed=FALSE,
## catsize=TRUE} => {type=reptile} 0.01111111 0.5000000 9.00000 0.2222222
## [13] {eggs=FALSE,
## venomous=TRUE} => {type=mollusc.et.al} 0.01111111 0.5000000 5.00000 0.2222222
##
## Confusion Matrix and Statistics
##
## Reference
## Prediction mammal bird reptile fish amphibian insect mollusc.et.al
## mammal 5 0 0 0 0 0 0
## bird 0 0 0 0 0 0 0
## reptile 0 0 0 0 1 0 0
## fish 0 0 0 3 0 0 0
## amphibian 0 0 0 0 0 0 0
## insect 0 0 0 0 0 1 0
## mollusc.et.al 0 0 0 0 0 0 1
##
## Overall Statistics
##
## Accuracy : 0.9091
## 95% CI : (0.5872, 0.9977)
## No Information Rate : 0.4545
## P-Value [Acc > NIR] : 0.00243
##
## Kappa : 0.8706
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: mammal Class: bird Class: reptile Class: fish
## Sensitivity 1.0000 NA NA 1.0000
## Specificity 1.0000 1 0.90909 1.0000
## Pos Pred Value 1.0000 NA NA 1.0000
## Neg Pred Value 1.0000 NA NA 1.0000
## Prevalence 0.4545 0 0.00000 0.2727
## Detection Rate 0.4545 0 0.00000 0.2727
## Detection Prevalence 0.4545 0 0.09091 0.2727
## Balanced Accuracy 1.0000 NA NA 1.0000
## Class: amphibian Class: insect Class: mollusc.et.al
## Sensitivity 0.00000 1.00000 1.00000
## Specificity 1.00000 1.00000 1.00000
## Pos Pred Value NaN 1.00000 1.00000
## Neg Pred Value 0.90909 1.00000 1.00000
## Prevalence 0.09091 0.09091 0.09091
## Detection Rate 0.00000 0.09091 0.09091
## Detection Prevalence 0.00000 0.09091 0.09091
## Balanced Accuracy 0.50000 1.00000 1.00000
##
## ---------------------------------------------------------------------
##
## CBA Classifier Object
## Formula: type ~ .
## Number of rules: 7
## Default Class: mammal
## Classification method: first
## Description: RWeka classifier weka/classifiers/rules/JRip
##
##
## Rule base:
## lhs rhs support confidence
## [1] {feathers=FALSE,
## aquatic=TRUE,
## breathes=TRUE,
## catsize=FALSE} => {type=amphibian} 0.03333333 1.0000000
## [2] {hair=FALSE,
## toothed=TRUE,
## fins=FALSE} => {type=reptile} 0.04444444 0.5714286
## [3] {predator=FALSE,
## backbone=FALSE,
## legs=TRUE} => {type=insect} 0.06666667 1.0000000
## [4] {backbone=FALSE} => {type=mollusc.et.al} 0.10000000 0.5625000
## [5] {breathes=FALSE} => {type=fish} 0.11111111 0.5882353
## [6] {feathers=TRUE} => {type=bird} 0.22222222 1.0000000
## [7] {} => {type=mammal} 0.40000000 0.4000000
##
## Confusion Matrix and Statistics
##
## Reference
## Prediction mammal bird reptile fish amphibian insect mollusc.et.al
## mammal 5 0 0 0 0 0 0
## bird 0 0 0 0 0 0 0
## reptile 0 0 0 0 0 0 0
## fish 0 0 0 3 0 0 0
## amphibian 0 0 0 0 1 0 0
## insect 0 0 0 0 0 1 0
## mollusc.et.al 0 0 0 0 0 0 1
##
## Overall Statistics
##
## Accuracy : 1
## 95% CI : (0.7151, 1)
## No Information Rate : 0.4545
## P-Value [Acc > NIR] : 0.0001711
##
## Kappa : 1
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: mammal Class: bird Class: reptile Class: fish
## Sensitivity 1.0000 NA NA 1.0000
## Specificity 1.0000 1 1 1.0000
## Pos Pred Value 1.0000 NA NA 1.0000
## Neg Pred Value 1.0000 NA NA 1.0000
## Prevalence 0.4545 0 0 0.2727
## Detection Rate 0.4545 0 0 0.2727
## Detection Prevalence 0.4545 0 0 0.2727
## Balanced Accuracy 1.0000 NA NA 1.0000
## Class: amphibian Class: insect Class: mollusc.et.al
## Sensitivity 1.00000 1.00000 1.00000
## Specificity 1.00000 1.00000 1.00000
## Pos Pred Value 1.00000 1.00000 1.00000
## Neg Pred Value 1.00000 1.00000 1.00000
## Prevalence 0.09091 0.09091 0.09091
## Detection Rate 0.09091 0.09091 0.09091
## Detection Prevalence 0.09091 0.09091 0.09091
## Balanced Accuracy 1.00000 1.00000 1.00000
##
## ---------------------------------------------------------------------
##
## CBA Classifier Object
## Formula: type ~ .
## Number of rules: 8
## Default Class: mammal
## Classification method: first
## Description: RWeka classifier weka/classifiers/rules/PART
##
##
## Rule base:
## lhs rhs support confidence
## [1] {feathers=FALSE,
## milk=TRUE} => {type=mammal} 0.40000000 1.00000000
## [2] {feathers=TRUE} => {type=bird} 0.22222222 1.00000000
## [3] {airborne=FALSE,
## predator=TRUE,
## backbone=FALSE} => {type=mollusc.et.al} 0.07777778 1.00000000
## [4] {backbone=FALSE,
## legs=TRUE} => {type=insect} 0.07777778 0.53846154
## [5] {fins=TRUE} => {type=fish} 0.11111111 0.71428571
## [6] {backbone=TRUE,
## tail=TRUE} => {type=reptile} 0.05555556 0.07575758
## [7] {aquatic=TRUE} => {type=amphibian} 0.03333333 0.09677419
## [8] {} => {type=mollusc.et.al} 0.10000000 0.10000000
##
## Confusion Matrix and Statistics
##
## Reference
## Prediction mammal bird reptile fish amphibian insect mollusc.et.al
## mammal 5 0 0 0 0 0 0
## bird 0 0 0 0 0 0 0
## reptile 0 0 0 0 0 0 0
## fish 0 0 0 3 0 0 0
## amphibian 0 0 0 0 1 0 0
## insect 0 0 0 0 0 1 0
## mollusc.et.al 0 0 0 0 0 0 1
##
## Overall Statistics
##
## Accuracy : 1
## 95% CI : (0.7151, 1)
## No Information Rate : 0.4545
## P-Value [Acc > NIR] : 0.0001711
##
## Kappa : 1
##
## Mcnemar's Test P-Value : NA
##
## Statistics by Class:
##
## Class: mammal Class: bird Class: reptile Class: fish
## Sensitivity 1.0000 NA NA 1.0000
## Specificity 1.0000 1 1 1.0000
## Pos Pred Value 1.0000 NA NA 1.0000
## Neg Pred Value 1.0000 NA NA 1.0000
## Prevalence 0.4545 0 0 0.2727
## Detection Rate 0.4545 0 0 0.2727
## Detection Prevalence 0.4545 0 0 0.2727
## Balanced Accuracy 1.0000 NA NA 1.0000
## Class: amphibian Class: insect Class: mollusc.et.al
## Sensitivity 1.00000 1.00000 1.00000
## Specificity 1.00000 1.00000 1.00000
## Pos Pred Value 1.00000 1.00000 1.00000
## Neg Pred Value 1.00000 1.00000 1.00000
## Prevalence 0.09091 0.09091 0.09091
## Detection Rate 0.09091 0.09091 0.09091
## Detection Prevalence 0.09091 0.09091 0.09091
## Balanced Accuracy 1.00000 1.00000 1.00000cbind(
accuracy = unlist(accuracy),
nrules = unlist(nrules)
)## accuracy nrules
## CBA algorithm (Liu et al., 1998).Accuracy 0.8181818 4
## RWeka classifier weka/classifiers/trees/J48.Accuracy 0.9090909 9
## FOIL-based classifier (Yin and Han, 2003).Accuracy 0.9090909 13
## RWeka classifier weka/classifiers/rules/JRip.Accuracy 1.0000000 7
## RWeka classifier weka/classifiers/rules/PART.Accuracy 1.0000000 8