rm(list=ls(all=TRUE))
knitr::opts_chunk$set(comment = NA)
knitr::opts_knit$set(global.par = TRUE)
options(scipen=20, digits=5, width=100, tibble.print_min=4)
rmarkdown::find_pandoc(version = '2.7.3')## $version
## [1] '0'
##
## $dir
## NULL
if(!require(heatmaply)) install.packages("heatmaply")
if(!require(Matrix)) install.packages("Matrix")
if(!require(arules)) install.packages("arules")
pacman::p_load(ggplot2,dplyr,heatmaply)
pacman::p_load(Matrix, arules) cust tid items
32241 119328 817182
[1] 119422
[1] 23789
pk = p$prod[1:2000]
Z = filter(Z0, prod %in% pk)
L = split(Z$prod, Z$tid)
tr = as(L, "transactions"); trtransactions in sparse format with
107797 transactions (rows) and
2000 items (columns)
Apriori
Parameter specification:
confidence minval smax arem aval originalSupport maxtime support minlen maxlen target ext
0.5 0.1 1 none FALSE TRUE 5 0.0002 1 10 rules TRUE
Algorithmic control:
filter tree heap memopt load sort verbose
0.1 TRUE TRUE FALSE TRUE 2 TRUE
Absolute minimum support count: 21
set item appearances ...[0 item(s)] done [0.00s].
set transactions ...[2000 item(s), 107797 transaction(s)] done [0.13s].
sorting and recoding items ... [2000 item(s)] done [0.00s].
creating transaction tree ... done [0.03s].
checking subsets of size 1 2 3 4 5 6 done [0.04s].
writing ... [1001 rule(s)] done [0.01s].
creating S4 object ... done [0.01s].
set of 1001 rules
rule length distribution (lhs + rhs):sizes
2 3 4 5 6
85 558 289 64 5
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.00 3.00 3.00 3.35 4.00 6.00
summary of quality measures:
support confidence coverage lift count
Min. :0.000204 Min. :0.500 Min. :0.000232 Min. : 6.64 Min. : 22
1st Qu.:0.000232 1st Qu.:0.557 1st Qu.:0.000371 1st Qu.: 60.48 1st Qu.: 25
Median :0.000297 Median :0.636 Median :0.000482 Median : 91.25 Median : 32
Mean :0.000519 Mean :0.652 Mean :0.000829 Mean : 211.00 Mean : 56
3rd Qu.:0.000501 3rd Qu.:0.736 3rd Qu.:0.000789 3rd Qu.: 281.70 3rd Qu.: 54
Max. :0.006494 Max. :0.966 Max. :0.009648 Max. :1131.30 Max. :700
mining info:
data ntransactions support confidence
tr 107797 0.0002 0.5
call
apriori(data = tr, parameter = list(supp = 0.0002, conf = 0.5))
lhs rhs support confidence coverage lift count
[1] {4710030346110} => {4710030346097} 0.00094622 0.55135 0.0017162 264.15 102
[2] {4710030346110} => {4710030346103} 0.00111320 0.64865 0.0017162 272.07 120
[3] {4710030346110} => {4710030346059} 0.00105754 0.61622 0.0017162 202.52 114
[4] {4716114000312} => {4716114000329} 0.00141006 0.55273 0.0025511 209.06 152
[5] {4716114000329} => {4716114000312} 0.00141006 0.53333 0.0026439 209.06 152
[6] {4710030346097} => {4710030346103} 0.00109465 0.52444 0.0020873 219.97 118
[7] {4710030346097} => {4710030346059} 0.00121525 0.58222 0.0020873 191.35 131
[8] {4719859015061} => {4713080626225} 0.00105754 0.50667 0.0020873 241.67 114
[9] {4713080626225} => {4719859015061} 0.00105754 0.50442 0.0020965 241.67 114
[10] {4711524000617} => {4711524000419} 0.00094622 0.54545 0.0017347 241.97 102
[11] {4710030346103} => {4710030346059} 0.00155848 0.65370 0.0023841 214.84 168
[12] {4710030346059} => {4710030346103} 0.00155848 0.51220 0.0030428 214.84 168
[13] {4710008241119} => {4710008241218} 0.00094622 0.52041 0.0018182 177.53 102
[14] {4711524000457} => {4711524000419} 0.00095550 0.50490 0.0018924 223.98 103
[15] {4711524000457} => {4711524000396} 0.00101116 0.53431 0.0018924 155.67 109
[16] {0076150430530} => {0076150215281} 0.00108537 0.51770 0.0020965 169.11 117
[17] {4719090701051} => {4719090790000} 0.00163270 0.57705 0.0028294 110.10 176
[18] {4711524000471} => {4711524000396} 0.00108537 0.51092 0.0021244 148.85 117
[19] {4711524000495} => {4711524000396} 0.00115031 0.50000 0.0023006 145.67 124
[20] {4711524000419} => {4711524000396} 0.00153065 0.67901 0.0022542 197.83 165
[21] {4710321861209} => {4710321861186} 0.00207798 0.60870 0.0034138 124.51 224
[22] {4710321871260} => {4710321861186} 0.00185534 0.54054 0.0034324 110.57 200
[23] {4711524000907} => {4711524000891} 0.00143789 0.56364 0.0025511 172.61 155
[24] {4711524000907} => {4711524001041} 0.00166053 0.65091 0.0025511 175.42 179
[25] {0719859796124} => {0719859796117} 0.00232845 0.69529 0.0033489 132.89 251
[26] {4719090790017} => {4719090790000} 0.00325612 0.80876 0.0040261 154.30 351
[27] {4719090790000} => {4719090790017} 0.00325612 0.62124 0.0052413 154.30 351
[28] {4711524000891} => {4711524001041} 0.00195738 0.59943 0.0032654 161.54 211
[29] {4711524001041} => {4711524000891} 0.00195738 0.52750 0.0037107 161.54 211
[30] {4719090701051, 4719090790017} => {4719090790000} 0.00102971 0.84733 0.0012152 161.66 111
[31] {4719090701051, 4719090790000} => {4719090790017} 0.00102971 0.63068 0.0016327 156.65 111
[32] {4710321861209, 4710321871260} => {4710321861186} 0.00102044 0.63584 0.0016049 130.06 110
[33] {4710321861186, 4710321871260} => {4710321861209} 0.00102044 0.55000 0.0018553 161.11 110
[34] {4711524000891, 4711524000907} => {4711524001041} 0.00110393 0.76774 0.0014379 206.90 119
[35] {4711524000907, 4711524001041} => {4711524000891} 0.00110393 0.66480 0.0016605 203.59 119
[36] {4711524000891, 4711524001041} => {4711524000907} 0.00110393 0.56398 0.0019574 221.07 119
☝️ The Association Rules …
lhs item(s) is in a cartrhs item(s) tend to be also in that cartsupport := the ratio that this rule actually
occurredconfidence := the probability that the rhs
are also in cartcount := the number of occurrences of this rule