arules/vignettes/preparing-transaction-data.Rmd at master · mhahsler/arules · GitHub

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
---
title: "Preparing transaction data"
author: "Michael Hahsler"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Preparing transaction data}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
library(arules)
set.seed(1234)
```

Data for association rule mining comes from many sources and in several
layouts. `arules` stores these data in the sparse `transactions` class, and
the `transactions()` constructor accepts several common input layouts.

The following examples show how to convert each layout. Always inspect the
resulting `transactions` object with `summary()` or `itemLabels()`: values that
were encoded incorrectly in the source data may otherwise become unintended
items.

## A list of baskets

Use one character vector per transaction. List names become transaction IDs.

```{r}
baskets <- list(
  order_1 = c("apple", "bread"),
  order_2 = c("bread", "milk"),
  order_3 = c("apple", "bread", "milk")
)
from_list <- transactions(baskets)
inspect(from_list)
```

Check both the transaction summary and the resulting item labels.

```{r}
summary(from_list)
itemLabels(from_list)
```

The item labels confirm that the baskets were translated correctly.

## A binary matrix

Rows represent transactions and columns represent items. Logical matrices make
the intended coding explicit.

```{r}
binary <- matrix(
  c(TRUE, TRUE, FALSE,
    FALSE, TRUE, TRUE,
    TRUE, TRUE, TRUE),
  nrow = 3,
  byrow = TRUE,
  dimnames = list(names(baskets), c("apple", "bread", "milk"))
)
from_matrix <- transactions(binary)

itemLabels(from_matrix)
inspect(from_matrix)
```

## A data frame in wide format

Categorical columns are converted to items of the form `variable=value`.
Logical columns represent the presence or absence of a single item. Missing
values are omitted.

```{r}
customers <- data.frame(
  age_group = factor(c("young", "adult", "adult")),
  region = factor(c("north", "south", "north")),
  subscriber = c(TRUE, FALSE, TRUE)
)
from_wide <- transactions(customers)

itemLabels(from_wide)
inspect(from_wide)
```

Continuous variables need to be discretized before conversion.

```{r}
measurements <- data.frame(
  spend = c(12, 18, 35, 42, 55),
  visits = c(1, 2, 3, 5, 8)
)
measurements_discrete <- discretizeDF(
  measurements,
  default = list(method = "frequency", breaks = 2)
)
from_discrete <- transactions(measurements_discrete)

itemLabels(from_discrete)
inspect(from_discrete)
```

## A data frame in long format

Long-format data has one row per transaction--item pair. Identify the
transaction and item columns with `cols`.

```{r}
long <- data.frame(
  order = c(1, 1, 2, 2, 3),
  product = c("apple", "bread", "bread", "milk", "apple")
)
from_long <- transactions(long, format = "long", cols = c("order", "product"))

itemLabels(from_long)
inspect(from_long)
```