itemMatrix-class: Class itemMatrix --- Sparse Binary Incidence Matrix to Represent Sets of Items

Description

The itemMatrix class is the basic building block for transactions, itemsets and rules in package arules. The class contains a sparse Matrix representation of items (a set of itemsets or transactions) and the corresponding item labels.

Arguments

Objects from the Class

Objects can be created by calls of the form new("itemMatrix", ...). However, most of the time objects will be created by coercion from a matrix, list or data.frame.

Slots

data:: Object of class ngCMatrix (from package Matrix) which stores item occurrences in sparse representation. Note that the ngCMatrix is column-oriented and itemMatrix is row-oriented with each row representing an element (an itemset, a transaction, etc.). As a result, the ngCMatrix in this slot is always a transposed version of the binary incidence matrix in itemMatrix.
itemInfo:: a data.frame which contains named vectors of the length equal to the number of elements in the set. If the slot is not empty (contains no item labels), the first element in the data.frame must have the name "labels" and contain a character vector with the item labels used for representing an item. In addition to the item labels, the data.frame can contain arbitrary named vectors (of the same length) to represent, e.g., variable names and values which were used to create the binary items or hierarchical category information associated with each item label.
itemsetInfo:: a data.frame which may contain additional information for the rows (mostly representing itemsets) in the matrix.

Methods

coerce: signature(from = "matrix", to = "itemMatrix"); expects from to be a binary matrix only containing 0s and 1s.
coerce: signature(from = "itemMatrix", to = "matrix"); coerces to a dense 0-1 matrix of storage.mode "integer" instead of "double" to save memory.
coerce: signature(from = "list", to = "itemMatrix"); from is a list of vectors. Each vector contains one set/transaction/….
coerce: signature(from = "itemMatrix", to = "list"); see also the methods for LIST.
coerce: signature(from = "itemMatrix", to = "ngCMatrix"); access the sparse matrix representation. Note, the ngCMatrix contains a transposed from of the itemMatrix.
coerce: signature(from = "ngCMatrix", to = "itemMatrix"); Note, the ngCMatrix has to be transposed with items as rows!
c: signature(object = "itemMatrix"); combine.
dim: signature(x = "itemMatrix"); returns the dimensions of the itemMatrix.
dimnames, rownames, colnames: signature(x = "itemMatrix"); returns row (itemsetID) and column (item) names.
dimnames: signature(x = "itemMatrix"); returns dimnames.
dimnames<-: signature(x = "itemMatrix", value = "list"); replace dimnames.
%in%: signature(x = "itemMatrix", table = "character"); matches the strings in table against the item labels in x and returns a logical vector indicating if a row (itemset) in x contains any of the items specified in table. Note that there is a %in% method with signature(x = "itemMatrix", table = "character"). This method is described in together with match.
%ain%: signature(x = "itemMatrix", table = "character"); matches the strings in table against the item labels in x and returns a logical vector indicating if a row (itemset) in x contains all of the items specified in table.
%oin%: signature(x = "itemMatrix", table = "character"); matches the strings in table against the item labels in x and returns a logical vector indicating if a row (itemset) in x contains only items specified in table.
%pin%: signature(x = "itemMatrix", table = "character"); matches the strings in table against the item labels in x (using partial matching) and returns a logical vector indicating if a row (itemset) in x contains any of the items specified in table.
itemLabels: signature(object = "itemMatrix"); returns the item labels used for encoding as a character vector.
itemLabels<-: signature(object = "itemMatrix"); replaces the item labels used for encoding.
itemInfo: signature(object = "itemMatrix"); returns the whole item/column information data.frame including labels.
itemInfo<-: signature(object = "itemMatrix"); replaces the item/column info by a data.frame.
itemsetInfo: signature(object = "itemMatrix"); returns the item set/row information data.frame.
itemsetInfo<-: signature(object = "itemMatrix"); replaces the item set/row info by a data.frame.
labels: signature(x = "transactions"); returns labels for the itemsets. The following arguments can be used to customize the representation of the labels: itemSep, setStart and setEnd.
nitems: signature(x = "itemMatrix"); returns the number of items (number in columns) in the itemMatrix.
show: signature(object = "itemMatrix")
summary: signature(object = "itemMatrix")

Details

Sets of itemsets are represented as sparse binary matrices. If you work with several itemMatrices at the same time (e.g., several transaction sets, lhs and rhs of a rule, etc.), then the encoding (itemLabes and order of the items in the binary matrix) in the different itemMatrices is important and needs to conform. See itemCoding to learn how to encode and recode itemMatrix objects.

Examples

Run this code

# NOT RUN {
set.seed(1234)
  
## Generate random data and coerce data to itemMatrix.
m <- matrix(runif(100000)>0.8, ncol=20)
dimnames(m) <- list(NULL, paste("item", c(1:20), sep=""))
i <- as(m, "itemMatrix")

## Get the number of elements (rows) in the itemMatrix.
length(i)

## Get first 5 elements (rows) of the itemMatrix as list.
as(i[1:5], "list")

## Get first 5 elements (rows) of the itemMatrix as matrix.
as(i[1:5], "matrix")

## Get first 5 elements (rows) of the itemMatrix as sparse ngCMatrix.
## Warning: for efficiency reasons, the ngCMatrix you get is transposed!
as(i[1:5], "ngCMatrix")

## Get labels for the first 5 itemsets (first default and then with 
## custom formating)
labels(i[1:5])
labels(i[1:5], itemSep = " + ", setStart = "", setEnd = "")

## create itemsets from itemMatrix  
is <- new("itemsets", items = i[1:3])
inspect(is)

## create rules (rhs and lhs cannot share items so I use 
## itemSetdiff here). Also assign (random) support.
rules <- new("rules", lhs=itemSetdiff(i[4:6],i[1:3]), rhs=i[1:3],
  quality = data.frame(support = runif(3)))
inspect(rules) 
# }

Run the code above in your browser using DataCamp Workspace