| title | author | date | output | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
Market Basket Analysis |
Jennifer Brosnahan |
7/20/2020 |
|
We have been asked by Blackwell Electronics to conduct a Market Basket Analysis on 30 days worth of transactions. Based on this market basket analysis, we will provide insights and purchasing patterns in order to provide product recommendations for cross-selling and recommender opportunities.
The purpose of this project is to conduct a Market Basket Analysis to gain insights on customer electronic transactions and find patterns or product relationships. Not only will we apply rules and gain insights, but we will also provide key business insights related to our findings. The following questions will be addressed:
- Are there any interesting patterns or item relationships within Blackwell Electronics transactions?
- Based on your market basket analysis, what product recommendations do you have for Blackwell?
The dataset contains 30 days' worth of Blackwell Electronics online transactions. Each row is a transaction, which contains only names of items purchased. There are no other variables, numeric or otherwise.
We will use basket formatting and apply the apriori algorithm to determine relationships between products.
library(tidyverse)
library(openxlsx)
library(knitr)
library(ggplot2)
library(arules)
library(arulesViz)
library(dplyr)
library(kableExtra)trans <- read.transactions(file.path('C:/Users/jlbro/OneDrive/C3T4', 'ElectronidexTransactions.csv'), format = 'basket', sep=',', rm.duplicates=TRUE)## distribution of transactions with duplicates:
## items
## 1 2
## 191 10
## summary statistics
summary(trans) ## transactions as itemMatrix in sparse format with
## 9835 rows (elements/itemsets/transactions) and
## 125 columns (items) and a density of 0.03506172
##
## most frequent items:
## iMac HP Laptop CYBERPOWER Gamer Desktop
## 2519 1909 1809
## Apple Earpods Apple MacBook Air (Other)
## 1715 1530 33622
##
## element (itemset/transaction) length distribution:
## sizes
## 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
## 2 2163 1647 1294 1021 856 646 540 439 353 247 171 119 77 72 56
## 16 17 18 19 20 21 22 23 25 26 27 29 30
## 41 26 20 10 10 10 5 3 1 1 3 1 1
##
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 2.000 3.000 4.383 6.000 30.000
##
## includes extended item information - examples:
## labels
## 1 1TB Portable External Hard Drive
## 2 2TB Portable External Hard Drive
## 3 3-Button Mouse
# view head
inspect(trans[1:5]) ## items
## [1] {Acer Aspire,
## Belkin Mouse Pad,
## Brother Printer Toner,
## VGA Monitor Cable}
## [2] {Apple Wireless Keyboard,
## Dell Desktop,
## Lenovo Desktop Computer}
## [3] {iMac}
## [4] {Acer Desktop,
## Intel Desktop,
## Lenovo Desktop Computer,
## XIBERIA Gaming Headset}
## [5] {ASUS Desktop,
## Epson Black Ink,
## HP Laptop,
## iMac}
itemFrequency(trans, type = 'absolute') ## Plot top 20 most frequently purchased items
itemFrequencyPlot(trans, topN=20, type = 'absolute', col = 1:20)
image(sample(trans, 150))
Graph reveals clusters of products purchased, however, it is hard to view product numbers
| Logitech.Wireless.Keyboard | 22 |
|---|---|
| VGA Monitor Cable | 22 |
| Panasonic On-Ear Stereo Headphones | 23 |
| 1TB Portable External Hard Drive | 27 |
| Canon Ink | 27 |
| Logitech Stereo Headset | 30 |
| Ethernet Cable | 32 |
| Canon Office Printer | 35 |
| Gaming Mouse Professional | 35 |
| Audio Cable | 36 |
- 9835 transactions
- 125 total products
- 43,151 total items were purchased
- Top 5: iMac, HP Laptop, Cyberpower Gamer Desktop, Apple Earpods, Apple MacBook Air
- Items/transaction range is 0-30
- Customers purchase 1 item the most, however, 2163 transactions only included 1 item, indicating potential for cross-selling opportunities
- The average items purchased is 4.4
- The top 7 products stand out, after that, item frequency seems to level off
The apriori algorithm is used to uncover insights pertaining to transactional datasets based on item frequency. It assesses rules using two types of measurements: 1) Support, which measures itemsets of rules frequency, and 2) Confidence, which measures accuracy of the rules. A rule that measure high in both support and confidence are considered strong.
The aim is to apply apriori rules until we find the 'sweet spot' of a useful amount of support, confidence, and lifts, generally greater than 1 for stronger rules.
## Apriori #1
Rules1 <- apriori(trans, parameter = list(supp = 0.1, conf = 0.8, minlen = 3)) #Covers 10% of transactions (N=983), 80% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.8 0.1 1 none FALSE TRUE 5 0.1 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 983
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [10 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 done [0.00s].
## writing ... [0 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules1, by='lift')) ## Apriori #2
Rules2 <- apriori(trans, parameter = list(supp = 0.01, conf = 0.8, minlen = 3)) #Covers 1% of transactions (N=98), 80% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.8 0.1 1 none FALSE TRUE 5 0.01 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 98
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [82 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [0 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules2, by='lift')) ## Apriori #3
Rules3 <- apriori(trans, parameter = list(supp = 0.005, conf = 0.8, minlen = 3)) #Covers .05% transactions (N=49), 80% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.8 0.1 1 none FALSE TRUE 5 0.005 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 49
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [109 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 5 done [0.00s].
## writing ... [1 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules3, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## Dell Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.005287239 0.8125 0.006507372 4.185928 52
## Apriori #4
Rules4 <- apriori(trans, parameter = list(supp = 0.005, conf = 0.7, minlen = 3)) #Covers .05% transactions (N=49), 70% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.7 0.1 1 none FALSE TRUE 5 0.005 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 49
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.01s].
## sorting and recoding items ... [109 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 5 done [0.00s].
## writing ... [3 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules4, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## Dell Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.005287239 0.8125000 0.006507372 4.185928 52
## [2] {ASUS 2 Monitor,
## Dell Desktop,
## Lenovo Desktop Computer} => {iMac} 0.005185562 0.7391304 0.007015760 2.885807 51
## [3] {ASUS 2 Monitor,
## ASUS Monitor} => {iMac} 0.005083884 0.7142857 0.007117438 2.788805 50
## Apriori #5
Rules5 <- apriori(trans, parameter = list(supp = 0.005, conf = 0.6, minlen = 3)) #Covers .05% transactions (N=49), 60% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.6 0.1 1 none FALSE TRUE 5 0.005 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 49
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [109 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 5 done [0.00s].
## writing ... [28 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules5, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## Dell Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.005287239 0.8125000 0.006507372 4.185928 52
## [2] {Acer Aspire,
## iMac,
## ViewSonic Monitor} => {HP Laptop} 0.006202339 0.6630435 0.009354347 3.415942 61
## [3] {Acer Desktop,
## iMac,
## ViewSonic Monitor} => {HP Laptop} 0.006405694 0.6363636 0.010066090 3.278489 63
## [4] {Dell Desktop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {HP Laptop} 0.006202339 0.6224490 0.009964413 3.206802 61
## [5] {Computer Game,
## ViewSonic Monitor} => {HP Laptop} 0.007422471 0.6186441 0.011997966 3.187200 73
## [6] {Computer Game,
## Dell Desktop} => {HP Laptop} 0.005693950 0.6086957 0.009354347 3.135946 56
## [7] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.010777834 0.6022727 0.017895272 3.102856 106
## [8] {ASUS 2 Monitor,
## Dell Desktop,
## Lenovo Desktop Computer} => {iMac} 0.005185562 0.7391304 0.007015760 2.885807 51
## [9] {ASUS 2 Monitor,
## ASUS Monitor} => {iMac} 0.005083884 0.7142857 0.007117438 2.788805 50
## [10] {ASUS 2 Monitor,
## Microsoft Office Home and Student 2016} => {iMac} 0.005185562 0.6986301 0.007422471 2.727681 51
## [11] {Dell Desktop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.006914082 0.6938776 0.009964413 2.709125 68
## [12] {Apple Magic Keyboard,
## Dell Desktop,
## Lenovo Desktop Computer} => {iMac} 0.005287239 0.6842105 0.007727504 2.671382 52
## [13] {Apple Magic Keyboard,
## ASUS Monitor} => {iMac} 0.006812405 0.6700000 0.010167768 2.615899 67
## [14] {Acer Desktop,
## HP Laptop,
## ViewSonic Monitor} => {iMac} 0.006405694 0.6562500 0.009761057 2.562215 63
## [15] {Acer Desktop,
## ASUS 2 Monitor} => {iMac} 0.006405694 0.6428571 0.009964413 2.509925 63
## [16] {ASUS Monitor,
## ViewSonic Monitor} => {iMac} 0.008235892 0.6377953 0.012913066 2.490161 81
## [17] {ASUS Monitor,
## Dell Desktop} => {iMac} 0.007930859 0.6341463 0.012506355 2.475915 78
## [18] {Acer Desktop,
## HP Laptop,
## Lenovo Desktop Computer} => {iMac} 0.006304016 0.6326531 0.009964413 2.470085 62
## [19] {ASUS Monitor,
## Lenovo Desktop Computer} => {iMac} 0.009761057 0.6315789 0.015455008 2.465891 96
## [20] {ASUS 2 Monitor,
## Dell Desktop} => {iMac} 0.009049314 0.6312057 0.014336553 2.464433 89
## [21] {Acer Desktop,
## Apple Magic Keyboard} => {iMac} 0.006710727 0.6226415 0.010777834 2.430996 66
## [22] {ASUS Monitor,
## Microsoft Office Home and Student 2016} => {iMac} 0.005998983 0.6145833 0.009761057 2.399534 59
## [23] {Belkin Mouse Pad,
## Microsoft Office Home and Student 2016} => {iMac} 0.005490595 0.6136364 0.008947636 2.395837 54
## [24] {Apple MacBook Pro,
## ASUS Monitor} => {iMac} 0.005388917 0.6022727 0.008947636 2.351470 53
## [25] {HP Laptop,
## HP Monitor,
## Lenovo Desktop Computer} => {iMac} 0.005388917 0.6022727 0.008947636 2.351470 53
## [26] {HP Laptop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.008439248 0.6014493 0.014031520 2.348255 83
## [27] {Acer Desktop,
## ASUS Monitor} => {iMac} 0.005795628 0.6000000 0.009659380 2.342596 57
## [28] {Dell Desktop,
## Microsoft Office Home and Student 2016} => {iMac} 0.009456024 0.6000000 0.015760041 2.342596 93
## Apriori #6
Rules6 <- apriori(trans, parameter = list(supp = 0.006, conf = 0.6, minlen = 3)) #Covers .06% transactions (N=59), 60% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.6 0.1 1 none FALSE TRUE 5 0.006 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 59
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [102 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 5 done [0.00s].
## writing ... [17 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules6, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## iMac,
## ViewSonic Monitor} => {HP Laptop} 0.006202339 0.6630435 0.009354347 3.415942 61
## [2] {Acer Desktop,
## iMac,
## ViewSonic Monitor} => {HP Laptop} 0.006405694 0.6363636 0.010066090 3.278489 63
## [3] {Dell Desktop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {HP Laptop} 0.006202339 0.6224490 0.009964413 3.206802 61
## [4] {Computer Game,
## ViewSonic Monitor} => {HP Laptop} 0.007422471 0.6186441 0.011997966 3.187200 73
## [5] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.010777834 0.6022727 0.017895272 3.102856 106
## [6] {Dell Desktop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.006914082 0.6938776 0.009964413 2.709125 68
## [7] {Apple Magic Keyboard,
## ASUS Monitor} => {iMac} 0.006812405 0.6700000 0.010167768 2.615899 67
## [8] {Acer Desktop,
## HP Laptop,
## ViewSonic Monitor} => {iMac} 0.006405694 0.6562500 0.009761057 2.562215 63
## [9] {Acer Desktop,
## ASUS 2 Monitor} => {iMac} 0.006405694 0.6428571 0.009964413 2.509925 63
## [10] {ASUS Monitor,
## ViewSonic Monitor} => {iMac} 0.008235892 0.6377953 0.012913066 2.490161 81
## [11] {ASUS Monitor,
## Dell Desktop} => {iMac} 0.007930859 0.6341463 0.012506355 2.475915 78
## [12] {Acer Desktop,
## HP Laptop,
## Lenovo Desktop Computer} => {iMac} 0.006304016 0.6326531 0.009964413 2.470085 62
## [13] {ASUS Monitor,
## Lenovo Desktop Computer} => {iMac} 0.009761057 0.6315789 0.015455008 2.465891 96
## [14] {ASUS 2 Monitor,
## Dell Desktop} => {iMac} 0.009049314 0.6312057 0.014336553 2.464433 89
## [15] {Acer Desktop,
## Apple Magic Keyboard} => {iMac} 0.006710727 0.6226415 0.010777834 2.430996 66
## [16] {HP Laptop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.008439248 0.6014493 0.014031520 2.348255 83
## [17] {Dell Desktop,
## Microsoft Office Home and Student 2016} => {iMac} 0.009456024 0.6000000 0.015760041 2.342596 93
## Apriori #7
Rules7 <- apriori(trans, parameter = list(supp = 0.007, conf = 0.6, minlen = 3)) #Covers .07% transactions (N=68), 60% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.6 0.1 1 none FALSE TRUE 5 0.007 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 68
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [97 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [8 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules7, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Computer Game,
## ViewSonic Monitor} => {HP Laptop} 0.007422471 0.6186441 0.01199797 3.187200 73
## [2] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.010777834 0.6022727 0.01789527 3.102856 106
## [3] {ASUS Monitor,
## ViewSonic Monitor} => {iMac} 0.008235892 0.6377953 0.01291307 2.490161 81
## [4] {ASUS Monitor,
## Dell Desktop} => {iMac} 0.007930859 0.6341463 0.01250635 2.475915 78
## [5] {ASUS Monitor,
## Lenovo Desktop Computer} => {iMac} 0.009761057 0.6315789 0.01545501 2.465891 96
## [6] {ASUS 2 Monitor,
## Dell Desktop} => {iMac} 0.009049314 0.6312057 0.01433655 2.464433 89
## [7] {HP Laptop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.008439248 0.6014493 0.01403152 2.348255 83
## [8] {Dell Desktop,
## Microsoft Office Home and Student 2016} => {iMac} 0.009456024 0.6000000 0.01576004 2.342596 93
## Apriori #8
Rules8 <- apriori(trans, parameter = list(supp = 0.008, conf = 0.6, minlen = 3)) #Covers .08% transactions (N=78), 60% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.6 0.1 1 none FALSE TRUE 5 0.008 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 78
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [93 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [6 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules8, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.010777834 0.6022727 0.01789527 3.102856 106
## [2] {ASUS Monitor,
## ViewSonic Monitor} => {iMac} 0.008235892 0.6377953 0.01291307 2.490161 81
## [3] {ASUS Monitor,
## Lenovo Desktop Computer} => {iMac} 0.009761057 0.6315789 0.01545501 2.465891 96
## [4] {ASUS 2 Monitor,
## Dell Desktop} => {iMac} 0.009049314 0.6312057 0.01433655 2.464433 89
## [5] {HP Laptop,
## Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.008439248 0.6014493 0.01403152 2.348255 83
## [6] {Dell Desktop,
## Microsoft Office Home and Student 2016} => {iMac} 0.009456024 0.6000000 0.01576004 2.342596 93
## Apriori #9
Rules9 <- apriori(trans, parameter = list(supp = 0.009, conf = 0.6, minlen = 3)) #Covers .09% transactions (N=88), 60% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.6 0.1 1 none FALSE TRUE 5 0.009 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 88
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [87 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [4 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules9, by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.010777834 0.6022727 0.01789527 3.102856 106
## [2] {ASUS Monitor,
## Lenovo Desktop Computer} => {iMac} 0.009761057 0.6315789 0.01545501 2.465891 96
## [3] {ASUS 2 Monitor,
## Dell Desktop} => {iMac} 0.009049314 0.6312057 0.01433655 2.464433 89
## [4] {Dell Desktop,
## Microsoft Office Home and Student 2016} => {iMac} 0.009456024 0.6000000 0.01576004 2.342596 93
## Apriori #10
Rules10 <- apriori(trans, parameter = list(supp = 0.01, conf = 0.6, minlen = 3)) #Covers 1% transactions (N=98), 60% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.6 0.1 1 none FALSE TRUE 5 0.01 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 98
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [82 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [1 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules10, by='lift')) ## lhs rhs support confidence
## [1] {Acer Aspire,ViewSonic Monitor} => {HP Laptop} 0.01077783 0.6022727
## coverage lift count
## [1] 0.01789527 3.102856 106
## Apriori #11
Rules11 <- apriori(trans, parameter = list(supp = 0.01, conf = 0.5, minlen = 3)) ###Covers 1% transactions (N=98), 50% correct## Apriori
##
## Parameter specification:
## confidence minval smax arem aval originalSupport maxtime support minlen
## 0.5 0.1 1 none FALSE TRUE 5 0.01 3
## maxlen target ext
## 10 rules TRUE
##
## Algorithmic control:
## filter tree heap memopt load sort verbose
## 0.1 TRUE TRUE FALSE TRUE 2 TRUE
##
## Absolute minimum support count: 98
##
## set item appearances ...[0 item(s)] done [0.00s].
## set transactions ...[125 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [82 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [19 rule(s)] done [0.00s].
## creating S4 object ... done [0.00s].
inspect(sort(Rules11[1:10], by='lift')) ## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.01077783 0.6022727 0.01789527 3.102856 106
## [2] {ASUS 2 Monitor,
## Lenovo Desktop Computer} => {iMac} 0.01087951 0.5911602 0.01840366 2.308083 107
## [3] {Apple Magic Keyboard,
## Dell Desktop} => {iMac} 0.01016777 0.5847953 0.01738688 2.283232 100
## [4] {ASUS Monitor,
## HP Laptop} => {iMac} 0.01179461 0.5829146 0.02023386 2.275889 116
## [5] {ASUS 2 Monitor,
## HP Laptop} => {iMac} 0.01108287 0.5828877 0.01901373 2.275784 109
## [6] {HP Laptop,
## Microsoft Office Home and Student 2016} => {iMac} 0.01291307 0.5521739 0.02338587 2.155868 127
## [7] {Acer Desktop,
## ViewSonic Monitor} => {iMac} 0.01006609 0.5439560 0.01850534 2.123782 99
## [8] {Apple Magic Keyboard,
## Lenovo Desktop Computer} => {iMac} 0.01138790 0.5161290 0.02206406 2.015137 112
## [9] {Apple Magic Keyboard,
## HP Laptop} => {iMac} 0.01474326 0.5105634 0.02887646 1.993406 145
## [10] {HP Laptop,
## HP Monitor} => {iMac} 0.01057448 0.5024155 0.02104728 1.961594 104
Apriori 11 generated 19 total association rules, lift = 1.9-3.1. This appears to be our best rule, because it generated 19 total rules, has a good amount of support (1% frequency), confidence (50% accuracy), and lift of 1.5-3.1 (>1 indicates effectiveness). We will further inspect support, confidence, and lift to confirm.
# inspect support
support11 <- data.frame(inspect(sort(Rules11, by = 'support', decreasing = TRUE))) #Dell desktop and ViewSonic Monitor (#2) had greatest support at 1.5% (n=150 count)## lhs rhs support confidence coverage lift count
## [1] {HP Laptop,
## Lenovo Desktop Computer} => {iMac} 0.02308083 0.5000000 0.04616167 1.952164 227
## [2] {Dell Desktop,
## Lenovo Desktop Computer} => {iMac} 0.01860702 0.5069252 0.03670564 1.979202 183
## [3] {Acer Desktop,
## HP Laptop} => {iMac} 0.01596340 0.5114007 0.03121505 1.996675 157
## [4] {Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.01576004 0.5555556 0.02836807 2.169071 155
## [5] {Dell Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.01525165 0.5747126 0.02653787 2.960869 150
## [6] {Apple Magic Keyboard,
## HP Laptop} => {iMac} 0.01474326 0.5105634 0.02887646 1.993406 145
## [7] {Dell Desktop,
## ViewSonic Monitor} => {iMac} 0.01474326 0.5555556 0.02653787 2.169071 145
## [8] {HP Laptop,
## Microsoft Office Home and Student 2016} => {iMac} 0.01291307 0.5521739 0.02338587 2.155868 127
## [9] {CYBERPOWER Gamer Desktop,
## ViewSonic Monitor} => {iMac} 0.01281139 0.5271967 0.02430097 2.058348 126
## [10] {Acer Desktop,
## Lenovo Desktop Computer} => {iMac} 0.01230300 0.5307018 0.02318251 2.072033 121
## [11] {CYBERPOWER Gamer Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.01220132 0.5020921 0.02430097 2.586734 120
## [12] {ASUS Monitor,
## HP Laptop} => {iMac} 0.01179461 0.5829146 0.02023386 2.275889 116
## [13] {Apple Magic Keyboard,
## Lenovo Desktop Computer} => {iMac} 0.01138790 0.5161290 0.02206406 2.015137 112
## [14] {ASUS 2 Monitor,
## HP Laptop} => {iMac} 0.01108287 0.5828877 0.01901373 2.275784 109
## [15] {ASUS 2 Monitor,
## Lenovo Desktop Computer} => {iMac} 0.01087951 0.5911602 0.01840366 2.308083 107
## [16] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.01077783 0.6022727 0.01789527 3.102856 106
## [17] {HP Laptop,
## HP Monitor} => {iMac} 0.01057448 0.5024155 0.02104728 1.961594 104
## [18] {Apple Magic Keyboard,
## Dell Desktop} => {iMac} 0.01016777 0.5847953 0.01738688 2.283232 100
## [19] {Acer Desktop,
## ViewSonic Monitor} => {iMac} 0.01006609 0.5439560 0.01850534 2.123782 99
support11## data frame with 0 columns and 0 rows
# inspect confidence
confidence11 <- data.frame(inspect(sort(Rules11, by = 'confidence', decreasing = TRUE))) #Acer Aspire and ViewSonic Monitor (#1) had greatest confidence at 60% (although 2nd rule 57%)## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.01077783 0.6022727 0.01789527 3.102856 106
## [2] {ASUS 2 Monitor,
## Lenovo Desktop Computer} => {iMac} 0.01087951 0.5911602 0.01840366 2.308083 107
## [3] {Apple Magic Keyboard,
## Dell Desktop} => {iMac} 0.01016777 0.5847953 0.01738688 2.283232 100
## [4] {ASUS Monitor,
## HP Laptop} => {iMac} 0.01179461 0.5829146 0.02023386 2.275889 116
## [5] {ASUS 2 Monitor,
## HP Laptop} => {iMac} 0.01108287 0.5828877 0.01901373 2.275784 109
## [6] {Dell Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.01525165 0.5747126 0.02653787 2.960869 150
## [7] {Dell Desktop,
## ViewSonic Monitor} => {iMac} 0.01474326 0.5555556 0.02653787 2.169071 145
## [8] {Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.01576004 0.5555556 0.02836807 2.169071 155
## [9] {HP Laptop,
## Microsoft Office Home and Student 2016} => {iMac} 0.01291307 0.5521739 0.02338587 2.155868 127
## [10] {Acer Desktop,
## ViewSonic Monitor} => {iMac} 0.01006609 0.5439560 0.01850534 2.123782 99
## [11] {Acer Desktop,
## Lenovo Desktop Computer} => {iMac} 0.01230300 0.5307018 0.02318251 2.072033 121
## [12] {CYBERPOWER Gamer Desktop,
## ViewSonic Monitor} => {iMac} 0.01281139 0.5271967 0.02430097 2.058348 126
## [13] {Apple Magic Keyboard,
## Lenovo Desktop Computer} => {iMac} 0.01138790 0.5161290 0.02206406 2.015137 112
## [14] {Acer Desktop,
## HP Laptop} => {iMac} 0.01596340 0.5114007 0.03121505 1.996675 157
## [15] {Apple Magic Keyboard,
## HP Laptop} => {iMac} 0.01474326 0.5105634 0.02887646 1.993406 145
## [16] {Dell Desktop,
## Lenovo Desktop Computer} => {iMac} 0.01860702 0.5069252 0.03670564 1.979202 183
## [17] {HP Laptop,
## HP Monitor} => {iMac} 0.01057448 0.5024155 0.02104728 1.961594 104
## [18] {CYBERPOWER Gamer Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.01220132 0.5020921 0.02430097 2.586734 120
## [19] {HP Laptop,
## Lenovo Desktop Computer} => {iMac} 0.02308083 0.5000000 0.04616167 1.952164 227
confidence11## data frame with 0 columns and 0 rows
# inspect lift
lift11 <- data.frame(inspect(sort(Rules11, by = 'lift', decreasing = TRUE)))## lhs rhs support confidence coverage lift count
## [1] {Acer Aspire,
## ViewSonic Monitor} => {HP Laptop} 0.01077783 0.6022727 0.01789527 3.102856 106
## [2] {Dell Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.01525165 0.5747126 0.02653787 2.960869 150
## [3] {CYBERPOWER Gamer Desktop,
## ViewSonic Monitor} => {HP Laptop} 0.01220132 0.5020921 0.02430097 2.586734 120
## [4] {ASUS 2 Monitor,
## Lenovo Desktop Computer} => {iMac} 0.01087951 0.5911602 0.01840366 2.308083 107
## [5] {Apple Magic Keyboard,
## Dell Desktop} => {iMac} 0.01016777 0.5847953 0.01738688 2.283232 100
## [6] {ASUS Monitor,
## HP Laptop} => {iMac} 0.01179461 0.5829146 0.02023386 2.275889 116
## [7] {ASUS 2 Monitor,
## HP Laptop} => {iMac} 0.01108287 0.5828877 0.01901373 2.275784 109
## [8] {Dell Desktop,
## ViewSonic Monitor} => {iMac} 0.01474326 0.5555556 0.02653787 2.169071 145
## [9] {Lenovo Desktop Computer,
## ViewSonic Monitor} => {iMac} 0.01576004 0.5555556 0.02836807 2.169071 155
## [10] {HP Laptop,
## Microsoft Office Home and Student 2016} => {iMac} 0.01291307 0.5521739 0.02338587 2.155868 127
## [11] {Acer Desktop,
## ViewSonic Monitor} => {iMac} 0.01006609 0.5439560 0.01850534 2.123782 99
## [12] {Acer Desktop,
## Lenovo Desktop Computer} => {iMac} 0.01230300 0.5307018 0.02318251 2.072033 121
## [13] {CYBERPOWER Gamer Desktop,
## ViewSonic Monitor} => {iMac} 0.01281139 0.5271967 0.02430097 2.058348 126
## [14] {Apple Magic Keyboard,
## Lenovo Desktop Computer} => {iMac} 0.01138790 0.5161290 0.02206406 2.015137 112
## [15] {Acer Desktop,
## HP Laptop} => {iMac} 0.01596340 0.5114007 0.03121505 1.996675 157
## [16] {Apple Magic Keyboard,
## HP Laptop} => {iMac} 0.01474326 0.5105634 0.02887646 1.993406 145
## [17] {Dell Desktop,
## Lenovo Desktop Computer} => {iMac} 0.01860702 0.5069252 0.03670564 1.979202 183
## [18] {HP Laptop,
## HP Monitor} => {iMac} 0.01057448 0.5024155 0.02104728 1.961594 104
## [19] {HP Laptop,
## Lenovo Desktop Computer} => {iMac} 0.02308083 0.5000000 0.04616167 1.952164 227
lift11## data frame with 0 columns and 0 rows
# check quality for rules 11
head(quality(Rules11))## support confidence coverage lift count
## 1 0.01087951 0.5911602 0.01840366 2.308083 107
## 2 0.01108287 0.5828877 0.01901373 2.275784 109
## 3 0.01179461 0.5829146 0.02023386 2.275889 116
## 4 0.01291307 0.5521739 0.02338587 2.155868 127
## 5 0.01057448 0.5024155 0.02104728 1.961594 104
## 6 0.01016777 0.5847953 0.01738688 2.283232 100
# check for redundancy
inspect(Rules11[is.redundant(Rules11)]) #No redundant rulesplot(Rules11)
plot(Rules11, method = 'graph')
summary(Rules11)## set of 19 rules
##
## rule length distribution (lhs + rhs):sizes
## 3
## 19
##
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3 3 3 3 3 3
##
## summary of quality measures:
## support confidence coverage lift
## Min. :0.01007 Min. :0.5000 Min. :0.01739 Min. :1.952
## 1st Qu.:0.01098 1st Qu.:0.5110 1st Qu.:0.01962 1st Qu.:2.006
## Median :0.01230 Median :0.5440 Median :0.02339 Median :2.156
## Mean :0.01343 Mean :0.5439 Mean :0.02495 Mean :2.234
## 3rd Qu.:0.01500 3rd Qu.:0.5788 3rd Qu.:0.02745 3rd Qu.:2.280
## Max. :0.02308 Max. :0.6023 Max. :0.04616 Max. :3.103
## count
## Min. : 99.0
## 1st Qu.:108.0
## Median :121.0
## Mean :132.1
## 3rd Qu.:147.5
## Max. :227.0
##
## mining info:
## data ntransactions support confidence
## trans 9835 0.01 0.5
| Rule.# | Items.purchased.together | X3 | Likely.to.be.purchased | Support | Confidence | Coverage | Lift | Count |
|---|---|---|---|---|---|---|---|---|
| [1] | {Acer Aspire,ViewSonic Monitor} | => | {HP Laptop} | 0.0107778 | 0.6022727 | 0.0178953 | 3.102856 | 106 |
| [2] | {Dell Desktop,ViewSonic Monitor} | => | {HP Laptop} | 0.0152517 | 0.5747126 | 0.0265379 | 2.960869 | 150 |
| [3] | {CYBERPOWER Gamer Desktop,ViewSonic Monitor} | => | {HP Laptop} | 0.0122013 | 0.5020921 | 0.0243010 | 2.586734 | 120 |
| [4] | {ASUS 2 Monitor,Lenovo Desktop Computer} | => | {iMac} | 0.0108795 | 0.5911602 | 0.0184037 | 2.308083 | 107 |
| [5] | {Apple Magic Keyboard,Dell Desktop} | => | {iMac} | 0.0101678 | 0.5847953 | 0.0173869 | 2.283232 | 100 |
| [6] | {ASUS Monitor,HP Laptop} | => | {iMac} | 0.0117946 | 0.5829146 | 0.0202339 | 2.275889 | 116 |
| [7] | {ASUS 2 Monitor,HP Laptop} | => | {iMac} | 0.0110829 | 0.5828877 | 0.0190137 | 2.275784 | 109 |
| [8] | {Dell Desktop,ViewSonic Monitor} | => | {iMac} | 0.0147433 | 0.5555556 | 0.0265379 | 2.169071 | 145 |
| [9] | {Lenovo Desktop Computer,ViewSonic Monitor} | => | {iMac} | 0.0157600 | 0.5555556 | 0.0283681 | 2.169071 | 155 |
| [10] | {HP Laptop,Microsoft Office Home and Student 2016} | => | {iMac} | 0.0129131 | 0.5521739 | 0.0233859 | 2.155868 | 127 |
| [11] | {Acer Desktop,ViewSonic Monitor} | => | {iMac} | 0.0100661 | 0.5439560 | 0.0185053 | 2.123782 | 99 |
| [12] | {Acer Desktop,Lenovo Desktop Computer} | => | {iMac} | 0.0123030 | 0.5307018 | 0.0231825 | 2.072033 | 121 |
| [13] | {CYBERPOWER Gamer Desktop,ViewSonic Monitor} | => | {iMac} | 0.0128114 | 0.5271967 | 0.0243010 | 2.058348 | 126 |
| [14] | {Apple Magic Keyboard,Lenovo Desktop Computer} | => | {iMac} | 0.0113879 | 0.5161290 | 0.0220641 | 2.015137 | 112 |
| [15] | {Acer Desktop,HP Laptop} | => | {iMac} | 0.0159634 | 0.5114007 | 0.0312150 | 1.996675 | 157 |
| [16] | {Apple Magic Keyboard,HP Laptop} | => | {iMac} | 0.0147433 | 0.5105634 | 0.0288765 | 1.993406 | 145 |
| [17] | {Dell Desktop,Lenovo Desktop Computer} | => | {iMac} | 0.0186070 | 0.5069252 | 0.0367056 | 1.979202 | 183 |
| [18] | {HP Laptop,HP Monitor} | => | {iMac} | 0.0105745 | 0.5024155 | 0.0210473 | 1.961594 | 104 |
| [19] | {HP Laptop,Lenovo Desktop Computer} | => | {iMac} | 0.0230808 | 0.5000000 | 0.0461617 | 1.952164 | 227 |
1. Customers who purchase a ViewSonic Monitor with either an Acer Aspire laptop, Dell Desktop, or a CYBERPOWER Gamer Desktop also purchase an HP Laptop.
- These itemset combinations would serve as an excellent recommendation of items frequently bought together to further boost sales of higher cost items.
2. Monitors (ViewSonic, ASUS 2, ASUS, HP Monitor) are a popular product purchased alongside desktop computers, as seen in 10 of 19 rules.
- ViewSonic and ASUS 2 are top and would be excellent recommendation items with any desktop computer brand.
3. iMac Desktops are often purchased with 1 or more other desktop computer brands (Lenovo, Dell, Acer), as seen in 1406 total transactions.
- It is unclear if customers are buying different brands and keeping all, or buying to test out different brands, only to eventually return less desired desktops.
- In this situation, it would be prudent to first investigate returns of desktops purchased alongside other desktop computer brands listed above. If customers are more often keeping all desktops, then these higher sale transactions could serve as excellent recommendation opportunities of associated products frequently bought together.
1. Initiate recommender systems for the following products listed within these item sets frequently bought together:
- Acer Aspire, ViewSonic Monitor, HP Laptop
- Dell Desktop, ViewSonic Monitor, HP Laptop
- CYBERPOWER Gamer Desktop, ViewSonic Monitor, HP Laptop
- ASUS Monitor, HP Laptop, iMac
- ASUS 2 Monitor, HP Laptop, iMac
- HP Laptop, Microsoft Office Home and Student 2016, iMac
- CYBERPOWER Gamer Desktop, ViewSonic Monitor, iMac
- HP Laptop, HP Monitor, iMac