CBA classifier for R - provides implementations of a classifier based on the "Classification Based on Associations" (CBA). It can be used for building classification models from association rules. Rules are pruned in the order of precedence given by the sort criteria and a default rule is added. The final classifier labels provided instances. CBA was originally proposed by Liu, B. Hsu, W. and Ma, Y (1998). Integrating Classification and Association Rule Mining. Proceedings KDD-98, New York, 27-31 August. AAAI. pp80-86.
If you publish your research that uses rCBA, please cite:
@inproceedings{Kuchar:2015:EasyMiner,
author = {Stanislav Vojir and Vaclav Zeman and Jaroslav Kuchar and Tomas Kliegr},
title = {EasyMiner/R Preview: Towards a Web Interface for Association Rule Learning and Classification in R},
booktitle = {Proceedings of the RuleML 2015 Challenge, the Special Track on Rule-based Recommender Systems for the Web of Data, the Special Industry Track and the RuleML 2015 Doctoral Consortium hosted by the 9th International Web Rule Symposium (RuleML 2015), Berlin, Germany, August 2-5, 2015.},
year = {2015}
}The package is available in CRAN repository:
install.packages('rCBA',dependencies=TRUE, repos="http://cran.us.r-project.org")Prerequisites:
- Java 8
- R packages - devtools, rJava, R.utils
R dependencies installation:
install.packages(c("devtools","rJava"),dependencies=TRUE, repos="http://cran.us.r-project.org")Reconfiguration of Java in R:
sudo R CMD javareconfRecompile and reinstall rJava:
install.packages('rJava', type='source', dependencies=TRUE, repos="http://cran.us.r-project.org")rCBA installation:
library("devtools")
devtools::install_github("jaroslav-kuchar/rCBA")Vagrant virtual server:
vagrant upRstudio server:
- http://localhost:8787/
- vagrant:vagrant
Example 1 - automatically build model (including pruning) + classification:
library("rCBA")
data("iris")
output <- rCBA::build(iris)
model <- output$model
predictions <- rCBA::classification(iris, model)
table(predictions)
sum(iris$Species==predictions, na.rm=TRUE) / length(predictions)Example 2 - apriori + pruning:
library("arules")
library("rCBA")
train <- read.csv("./train.csv",header=TRUE) # read data
txns <- as(train,"transactions") # convert
rules <- apriori(txns, parameter = list(confidence = 0.1, support= 0.1, minlen=1, maxlen=5)) # rule mining
rules <- subset( rules, subset = rhs %pin% "y=") # filter
rulesFrame <- as(rules,"data.frame") # convert
print(nrow(rulesFrame))
prunedRulesFrame <- rCBA::pruning(train, rulesFrame, method="m2cba") # m2cba(default)|m1cba|dcbrcba
print(nrow(prunedRulesFrame))Example 3 - apriori + classification:
library("arules")
library("rCBA")
data("iris")
train <- sapply(iris,as.factor)
train <- data.frame(train, check.names=FALSE)
txns <- as(train,"transactions")
rules = apriori(txns, parameter=list(support=0.03, confidence=0.03, minlen=2),
appearance = list(rhs=c("Species=setosa", "Species=versicolor", "Species=virginica"),default="lhs"))
rulesFrame <- as(rules,"data.frame")
predictions <- rCBA::classification(train,rulesFrame)
table(predictions)
sum(train$Species==predictions,na.rm=TRUE)/length(predictions)
prunedRulesFrame <- rCBA::pruning(train, rulesFrame, method="m2cba")
predictions <- rCBA::classification(train, prunedRulesFrame)
table(predictions)
sum(train$Species==predictions,na.rm=TRUE)/length(predictions)Example 4 - fp-growth + classification:
library("rCBA")
data("iris")
train <- sapply(iris,as.factor)
train <- data.frame(train, check.names=FALSE)
txns <- as(train,"transactions")
rules = fpgrowth(txns, support=0.03, confidence=0.03, maxlen=2, consequent="Species")
rulesFrame <- as(rules,"data.frame")
predictions <- rCBA::classification(train,rulesFrame)
table(predictions)
sum(train$Species==predictions,na.rm=TRUE)/length(predictions)
prunedRulesFrame <- rCBA::pruning(train, rulesFrame, method="m2cba")
predictions <- rCBA::classification(train, prunedRulesFrame)
table(predictions)
sum(train$Species==predictions,na.rm=TRUE)/length(predictions)Example 5 - fp-growth automatic build:
library("rCBA")
data("iris")
output <- rCBA::buildFPGrowth(iris, "Species")
model <- output$model
predictions <- rCBA::classification(iris, model)
table(predictions)
sum(iris$Species==predictions, na.rm=TRUE) / length(predictions)- Jaroslav Kuchař (https://github.com/jaroslav-kuchar)
- Tomáš Kliegr (https://github.com/kliegr)
Apache License Version 2.0
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent