library(arules)
library(arulesViz)
library(datasets)
library(knitr)

Analiza koszykowa

Zadanie 1.

Za pomocą analizy koszykowej zidentyfikować reguły pozwalające na zarządzanie przestrzenią w sklepie wielkopowierzchniowym.

Polecenie 1.

Dokonać przeglądu produktów na podstawie obiektu data frame „baza”. Wydzielić grupy produktów spożywczych, nabiałowych, pieczywa, mięsa, serów, warzyw, środków higieny i czystości. Opisać je, tak jak zostało to przedstawione na przykładzie “środki czystości”.

setwd("C:/Users/majal/OneDrive - Politechnika Gdańska/Pulpit/PG/AG2 SEM3/big data")
baza1<- read.csv("C:/Users/majal/OneDrive - Politechnika Gdańska/Pulpit/PG/AG2 SEM3/big data/groceries.csv", sep=",",header = FALSE)
groceries <- read.transactions("groceries.csv", sep = ",")

Najczęściej występujące produkty:

itemFrequencyPlot(groceries, topN = 30)

Grupy produktów na podstawie analizy unikalnych wartości:

categories_names <- c("Nabial","Pieczywo","Mieso","Warzywa i owoce","Srodki czystosci","Alkohol","Napoje","Przekaski")
categories <- list(
  "Nabial" = c("whole milk", "yogurt", "butter", "cheese","whipped/sour cream","    
UHT-milk"),
  "Pieczywo" = c("bread", "rolls/buns", "pastry","brown bread","waffles"),
  "Mieso" = c("chicken", "beef", "pork","sausage","frankfurter"),
  "Warzywa i owoce" = c("root vegetables", "other vegetables", "potatoes","pip fruit","tropical fruit","citrus fruit"),
  "Srodki czystosci" = c("detergent", "cleaner"),
  "Alkohol"= c("canned beer","bottled beer","white wine"),
  "Napoje"=c("soda","bottled water","coffee","misc. beverages","beverages","    
fruit/vegetable juice"),
  "Przekaski"=c("chocolate","ice cream","salty snack","specialty chocolate","specialty bar")
)
kable(categories, format = "html", caption="Nabial, Pieczywo, Mieso, Warzywa i owoce, Srodki czystosci, Alkohol, Napoje, Przekaski", col.names=NULL)
Nabial, Pieczywo, Mieso, Warzywa i owoce, Srodki czystosci, Alkohol, Napoje, Przekaski
whole milk
yogurt
butter
cheese
whipped/sour cream
UHT-milk
bread
rolls/buns
pastry
brown bread
waffles
chicken
beef
pork
sausage
frankfurter
root vegetables
other vegetables
potatoes
pip fruit
tropical fruit
citrus fruit
detergent
cleaner
canned beer
bottled beer
white wine
soda
bottled water
coffee
misc. beverages
beverages
fruit/vegetable juice
chocolate
ice cream
salty snack
specialty chocolate
specialty bar

Polecenie 2.  

Dokonać przeglądu składu koszyków na podstawie obiektu data frame „baza”. Zaproponować wstępne reguły lub wstępną listę produktów po lewej lub prawej stronie reguły asocjacyjnej.

Przykładowe transakcje:

kable(head(baza1, 50), format = "markdown")
V1 V2 V3 V4
citrus fruit semi-finished bread margarine ready soups
tropical fruit yogurt coffee
whole milk
pip fruit yogurt cream cheese meat spreads
other vegetables whole milk condensed milk long life bakery product
whole milk butter yogurt rice
abrasive cleaner
rolls/buns
other vegetables UHT-milk rolls/buns bottled beer
liquor (appetizer)
potted plants
whole milk cereals
tropical fruit other vegetables white bread bottled water
chocolate
citrus fruit tropical fruit whole milk butter
curd yogurt flour bottled water
dishes
beef
frankfurter rolls/buns soda
chicken tropical fruit
butter sugar fruit/vegetable juice newspapers
fruit/vegetable juice
packaged fruit/vegetables
chocolate
specialty bar
other vegetables
butter milk pastry
whole milk
tropical fruit cream cheese processed cheese detergent
newspapers
tropical fruit root vegetables other vegetables frozen dessert
rolls/buns flour sweet spreads salty snack
waffles candy bathroom cleaner
bottled water canned beer
yogurt
sausage rolls/buns soda chocolate
other vegetables
brown bread soda fruit/vegetable juice canned beer
newspapers shopping bags
yogurt beverages bottled water specialty bar
hamburger meat other vegetables rolls/buns spices
bottled water hygiene articles napkins
root vegetables other vegetables whole milk beverages
sugar
pork berries other vegetables whole milk
whipped/sour cream artif. sweetener soda abrasive cleaner
beef grapes detergent
pastry soda
fruit/vegetable juice
canned beer

Wstępne reguły:

rules_initial <- list(
  "(whole milk) => (other vegetables, yogurt)",
  "(bread) => (butter, jam)",
  "(soda) => (pastry,rolls/buns)"
)
rules_initial
## [[1]]
## [1] "(whole milk) => (other vegetables, yogurt)"
## 
## [[2]]
## [1] "(bread) => (butter, jam)"
## 
## [[3]]
## [1] "(soda) => (pastry,rolls/buns)"

Polecenie 3.

Obliczyć wstępnie wszystkie sensowne reguły i je posortować po poziomie ufności:

Reguły posortowane po poziomie ufności:

rules <- apriori(groceries, parameter = list(support = 0.006, confidence = 0.25, minlen = 2))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.25    0.1    1 none FALSE            TRUE       5   0.006      2
##  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 ...[169 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 done [0.00s].
## writing ... [463 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules <- sort(rules, by = "confidence", decreasing = TRUE)
inscpect1<-DATAFRAME(rules[1:10])
kable(inscpect1, format = 'markdown')
LHS RHS support confidence coverage lift count
221 {butter,whipped/sour cream} {whole milk} 0.0067107 0.6600000 0.0101678 2.583008 66
229 {butter,yogurt} {whole milk} 0.0093543 0.6388889 0.0146416 2.500387 92
225 {butter,root vegetables} {whole milk} 0.0082359 0.6377953 0.0129131 2.496107 81
183 {curd,tropical fruit} {whole milk} 0.0065074 0.6336634 0.0102694 2.479936 64
222 {butter,tropical fruit} {whole milk} 0.0062023 0.6224490 0.0099644 2.436047 61
452 {other vegetables,tropical fruit,yogurt} {whole milk} 0.0076258 0.6198347 0.0123030 2.425816 75
239 {domestic eggs,tropical fruit} {whole milk} 0.0069141 0.6071429 0.0113879 2.376144 68
456 {other vegetables,root vegetables,yogurt} {whole milk} 0.0078292 0.6062992 0.0129131 2.372842 77
242 {domestic eggs,root vegetables} {whole milk} 0.0085409 0.5957447 0.0143366 2.331536 84
318 {citrus fruit,root vegetables} {other vegetables} 0.0103711 0.5862069 0.0176919 3.029608 102

“Wsparcie, , support” to prawdopodobieństwo wystąpienia produktu lub zestawu produktów (mleko i chleb), liczebność wybranych transakcji (mleko i chleb) do liczebności wszystkich transakcji.

“Poziom ufności, confidence”  to prawdopodobieństwo warunkowe, liczebność występowania wybranych transakcji (mleko i chleb) do liczebności (mleko).

“Coverage” - wsparcie dla lewej strony, badanie częstości występowania artykułów.

Usunięcie regół powtarzających się:

rules_redundant <- rules[!is.redundant(rules)]
rules_redundant_top50 <- head(rules_redundant, n=50, by="confidence", decreasing=TRUE)
inscpect1.1<-DATAFRAME(rules_redundant_top50)
kable(inscpect1.1,format="markdown")
LHS RHS support confidence coverage lift count
221 {butter,whipped/sour cream} {whole milk} 0.0067107 0.6600000 0.0101678 2.583008 66
229 {butter,yogurt} {whole milk} 0.0093543 0.6388889 0.0146416 2.500387 92
225 {butter,root vegetables} {whole milk} 0.0082359 0.6377953 0.0129131 2.496107 81
183 {curd,tropical fruit} {whole milk} 0.0065074 0.6336634 0.0102694 2.479936 64
222 {butter,tropical fruit} {whole milk} 0.0062023 0.6224490 0.0099644 2.436047 61
452 {other vegetables,tropical fruit,yogurt} {whole milk} 0.0076258 0.6198347 0.0123030 2.425816 75
239 {domestic eggs,tropical fruit} {whole milk} 0.0069141 0.6071429 0.0113879 2.376144 68
456 {other vegetables,root vegetables,yogurt} {whole milk} 0.0078292 0.6062992 0.0129131 2.372842 77
242 {domestic eggs,root vegetables} {whole milk} 0.0085409 0.5957447 0.0143366 2.331536 84
318 {citrus fruit,root vegetables} {other vegetables} 0.0103711 0.5862069 0.0176919 3.029608 102
155 {other vegetables,sugar} {whole milk} 0.0063040 0.5849057 0.0107778 2.289115 62
449 {root vegetables,tropical fruit,whole milk} {other vegetables} 0.0070158 0.5847458 0.0119980 3.022057 69
386 {root vegetables,tropical fruit} {other vegetables} 0.0123030 0.5845411 0.0210473 3.020999 121
187 {curd,yogurt} {whole milk} 0.0100661 0.5823529 0.0172852 2.279125 99
261 {citrus fruit,whipped/sour cream} {whole milk} 0.0063040 0.5794393 0.0108795 2.267722 62
290 {pip fruit,root vegetables} {whole milk} 0.0089476 0.5751634 0.0155567 2.250988 88
189 {curd,other vegetables} {whole milk} 0.0098627 0.5739645 0.0171835 2.246296 97
232 {butter,other vegetables} {whole milk} 0.0114896 0.5736041 0.0200305 2.244885 113
266 {tropical fruit,whipped/sour cream} {whole milk} 0.0079309 0.5735294 0.0138282 2.244593 78
448 {other vegetables,root vegetables,tropical fruit} {whole milk} 0.0070158 0.5702479 0.0123030 2.231750 69
184 {curd,root vegetables} {whole milk} 0.0062023 0.5700935 0.0108795 2.231146 61
389 {root vegetables,tropical fruit} {whole milk} 0.0119980 0.5700483 0.0210473 2.230969 118
264 {tropical fruit,whipped/sour cream} {other vegetables} 0.0078292 0.5661765 0.0138282 2.926088 77
416 {root vegetables,yogurt} {whole milk} 0.0145399 0.5629921 0.0258261 2.203354 143
203 {frankfurter,yogurt} {whole milk} 0.0062023 0.5545455 0.0111845 2.170296 61
271 {root vegetables,whipped/sour cream} {whole milk} 0.0094560 0.5535714 0.0170819 2.166484 93
247 {domestic eggs,other vegetables} {whole milk} 0.0123030 0.5525114 0.0222674 2.162336 121
251 {fruit/vegetable juice,root vegetables} {other vegetables} 0.0066090 0.5508475 0.0119980 2.846865 65
199 {pork,rolls/buns} {whole milk} 0.0062023 0.5495495 0.0112862 2.150744 61
152 {onions,whole milk} {other vegetables} 0.0066090 0.5462185 0.0120996 2.822942 65
171 {frozen vegetables,other vegetables} {whole milk} 0.0096594 0.5428571 0.0177936 2.124552 95
253 {fruit/vegetable juice,root vegetables} {whole milk} 0.0065074 0.5423729 0.0119980 2.122657 64
244 {domestic eggs,yogurt} {whole milk} 0.0077275 0.5390071 0.0143366 2.109485 76
457 {root vegetables,whole milk,yogurt} {other vegetables} 0.0078292 0.5384615 0.0145399 2.782853 77
217 {margarine,rolls/buns} {whole milk} 0.0079309 0.5379310 0.0147433 2.105273 78
167 {frozen vegetables,root vegetables} {whole milk} 0.0062023 0.5350877 0.0115913 2.094146 61
278 {rolls/buns,whipped/sour cream} {whole milk} 0.0078292 0.5347222 0.0146416 2.092715 77
157 {cream cheese,yogurt} {whole milk} 0.0066090 0.5327869 0.0124047 2.085141 65
294 {pip fruit,yogurt} {whole milk} 0.0095577 0.5310734 0.0179969 2.078435 94
165 {frozen vegetables,root vegetables} {other vegetables} 0.0061007 0.5263158 0.0115913 2.720082 60
275 {whipped/sour cream,yogurt} {whole milk} 0.0108795 0.5245098 0.0207422 2.052747 107
422 {rolls/buns,root vegetables} {whole milk} 0.0127097 0.5230126 0.0243010 2.046888 125
14 {baking powder} {whole milk} 0.0092527 0.5229885 0.0176919 2.046794 91
288 {pip fruit,root vegetables} {other vegetables} 0.0081342 0.5228758 0.0155567 2.702304 80
177 {beef,yogurt} {whole milk} 0.0061007 0.5217391 0.0116929 2.041904 60
337 {sausage,tropical fruit} {whole milk} 0.0072191 0.5182482 0.0139298 2.028242 71
297 {other vegetables,pip fruit} {whole milk} 0.0135231 0.5175097 0.0261312 2.025351 133
400 {tropical fruit,yogurt} {whole milk} 0.0151500 0.5173611 0.0292832 2.024770 149
303 {pastry,yogurt} {whole milk} 0.0091510 0.5172414 0.0176919 2.024301 90
320 {citrus fruit,root vegetables} {whole milk} 0.0091510 0.5172414 0.0176919 2.024301 90

Polecenie 4.

Zobaczyć, które produkty występują najczęściej i skonfrontować to z regułami z polecenia 3. Znaleźć nowe reguły lub potwierdzić stare, gdzie produkty najczęściej występujące w koszyku są po prawej stronie (tutaj założyć w procedurze apriori odpowiednie produkty po prawej stronie), zaproponować do praktycznego wdrożenia reguły odwrotne. Można również zaproponować wybrane produkty po lewej stronie lub po obydwu stronach.

Najczęściej występujące produkty:

itemFrequencyPlot(groceries, topN = 30)

Generowanie reguł z konkretnymi produktami prawej stronie:

rules_rhs <- apriori(groceries, parameter = list(supp = 0.005, conf = 0.01),
                     appearance = list(rhs = "canned beer", default = "lhs"))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.01    0.1    1 none FALSE            TRUE       5   0.005      1
##  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 ...[1 item(s)] done [0.00s].
## set transactions ...[169 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [120 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [9 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules_rhs <- sort(rules_rhs, by = "confidence", decreasing = TRUE)
inscpect2<-DATAFRAME(rules_rhs[1:5])
kable(inscpect2,format="markdown")
LHS RHS support confidence coverage lift count
2 {shopping bags} {canned beer} 0.0113879 0.1155831 0.0985257 1.4879052 112
5 {soda} {canned beer} 0.0138282 0.0793003 0.1743772 1.0208356 136
1 {} {canned beer} 0.0776817 0.0776817 1.0000000 1.0000000 764
4 {bottled water} {canned beer} 0.0080325 0.0726771 0.1105236 0.9355749 79
3 {sausage} {canned beer} 0.0063040 0.0670996 0.0939502 0.8637752 62 
rules_rhs1 <- apriori(groceries, parameter = list(supp = 0.005, conf = 0.25),
                     appearance = list(rhs = "rolls/buns", default = "lhs"))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.25    0.1    1 none FALSE            TRUE       5   0.005      1
##  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 ...[1 item(s)] done [0.00s].
## set transactions ...[169 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [120 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [55 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules_rhs1 <- sort(rules_rhs1, by = "confidence", decreasing = TRUE)
inscpect3<-DATAFRAME(rules_rhs1[1:5])
kable(inscpect3,format="markdown")
LHS RHS support confidence coverage lift count
35 {sausage,soda} {rolls/buns} 0.0096594 0.3974895 0.0243010 2.161033 95
33 {sausage,shopping bags} {rolls/buns} 0.0059990 0.3831169 0.0156584 2.082894 59
18 {frankfurter,other vegetables} {rolls/buns} 0.0055923 0.3395062 0.0164718 1.845795 55
24 {newspapers,yogurt} {rolls/buns} 0.0050839 0.3311258 0.0153533 1.800234 50
37 {other vegetables,sausage} {rolls/buns} 0.0088460 0.3283019 0.0269446 1.784881 87 
rules_rhs2 <- apriori(groceries, parameter = list(supp = 0.005, conf = 0.25),
                     appearance = list(rhs = "other vegetables", default = "lhs"))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.25    0.1    1 none FALSE            TRUE       5   0.005      1
##  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 ...[1 item(s)] done [0.00s].
## set transactions ...[169 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [120 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [178 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules_rhs2 <- sort(rules_rhs2, by = "confidence", decreasing = TRUE)
inscpect4<-DATAFRAME(rules_rhs2[1:5])
kable(inscpect4,format="markdown")
LHS RHS support confidence coverage lift count
173 {citrus fruit,root vegetables,whole milk} {other vegetables} 0.0057956 0.6333333 0.0091510 3.273165 57
171 {pip fruit,root vegetables,whole milk} {other vegetables} 0.0054906 0.6136364 0.0089476 3.171368 54
111 {pip fruit,whipped/sour cream} {other vegetables} 0.0055923 0.6043956 0.0092527 3.123610 55
54 {onions,root vegetables} {other vegetables} 0.0056940 0.6021505 0.0094560 3.112008 56
132 {citrus fruit,root vegetables} {other vegetables} 0.0103711 0.5862069 0.0176919 3.029608 102 
rules_rhs3 <- apriori(groceries, parameter = list(supp = 0.005, conf = 0.25),
                     appearance = list(rhs = "whole milk", default = "lhs"))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.25    0.1    1 none FALSE            TRUE       5   0.005      1
##  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 ...[1 item(s)] done [0.00s].
## set transactions ...[169 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [120 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 3 4 done [0.00s].
## writing ... [222 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules_rhs3 <- sort(rules_rhs3, by = "confidence", decreasing = TRUE)
inscpect5<-DATAFRAME(rules_rhs3[1:5])
kable(inscpect5,format="markdown")
LHS RHS support confidence coverage lift count
217 {root vegetables,tropical fruit,yogurt} {whole milk} 0.0056940 0.7000000 0.0081342 2.739554 56
214 {other vegetables,pip fruit,root vegetables} {whole milk} 0.0054906 0.6750000 0.0081342 2.641713 54
121 {butter,whipped/sour cream} {whole milk} 0.0067107 0.6600000 0.0101678 2.583008 66
150 {pip fruit,whipped/sour cream} {whole milk} 0.0059990 0.6483516 0.0092527 2.537421 59
126 {butter,yogurt} {whole milk} 0.0093543 0.6388889 0.0146416 2.500387 92

Generowanie reguł z konkretnymi produktami lewej stronie:

rules_lhs<-apriori(data=groceries, parameter=list(supp=0.001,conf = 0.15,minlen=2), appearance = list(default="rhs",lhs="yogurt"))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.15    0.1    1 none FALSE            TRUE       5   0.001      2
##  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: 9 
## 
## set item appearances ...[1 item(s)] done [0.00s].
## set transactions ...[169 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [157 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 done [0.00s].
## writing ... [8 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules_lhs<-sort(rules_lhs, decreasing=TRUE,by="confidence")
inscpect6<-DATAFRAME(rules_lhs[1:5])
kable(inscpect6,format="markdown")
LHS RHS support confidence coverage lift count
8 {yogurt} {whole milk} 0.0560244 0.4016035 0.1395018 1.571735 551
7 {yogurt} {other vegetables} 0.0434164 0.3112245 0.1395018 1.608457 427
6 {yogurt} {rolls/buns} 0.0343671 0.2463557 0.1395018 1.339363 338
3 {yogurt} {tropical fruit} 0.0292832 0.2099125 0.1395018 2.000475 288
5 {yogurt} {soda} 0.0273513 0.1960641 0.1395018 1.124368 269 
rules_lhs1<-apriori(data=groceries, parameter=list(supp=0.001,conf = 0.15,minlen=2), appearance = list(default="rhs",lhs="bottled water"))
## Apriori
## 
## Parameter specification:
##  confidence minval smax arem  aval originalSupport maxtime support minlen
##        0.15    0.1    1 none FALSE            TRUE       5   0.001      2
##  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: 9 
## 
## set item appearances ...[1 item(s)] done [0.00s].
## set transactions ...[169 item(s), 9835 transaction(s)] done [0.00s].
## sorting and recoding items ... [157 item(s)] done [0.00s].
## creating transaction tree ... done [0.00s].
## checking subsets of size 1 2 done [0.00s].
## writing ... [6 rule(s)] done [0.00s].
## creating S4 object  ... done [0.00s].
rules_lhs1<-sort(rules_lhs1, decreasing=TRUE,by="confidence")
inscpect7<-DATAFRAME(rules_lhs1[1:5])
kable(inscpect7,format="markdown")
LHS RHS support confidence coverage lift count
6 {bottled water} {whole milk} 0.0343671 0.3109476 0.1105236 1.216940 338
2 {bottled water} {soda} 0.0289781 0.2621895 0.1105236 1.503577 285
5 {bottled water} {other vegetables} 0.0248094 0.2244710 0.1105236 1.160101 244
4 {bottled water} {rolls/buns} 0.0241993 0.2189512 0.1105236 1.190373 238
3 {bottled water} {yogurt} 0.0229792 0.2079117 0.1105236 1.490387 226

Reguły przygotowane do wykorzystania przy zarządzaniu przestrzenią w sklepie:

{butter, whipped/sour cream} => {whole milk}
{other vegetables, pip fruit, root vegetables} => {whole milk}

Whole milk jest produktem, po który klienci siegają najczęściej. Z reguł wynika, że są jest ono kupowane głównie w towarzystwie warzyw i owoców oraz innych produktów nabiałowych.

{sausage, soda} => {rolls/buns}
{frankfurter, other vegetables} => {rolls/buns}

Rolls/buns najczęściej kupowane są produktami odzwięrzęcymi (po niecałe 100 wystąpień każdej reguły).

{citrus fruit, root vegetables} => {other vegetables}

Wybór other vegetables, może sugerować, że klient zakupi więcej, niż jedno warzywo/owoc.

{bottled water} => {whole milk}
{bottled water} => {soda}

Stawiając bottled water po lewej stronie, z reguł wychodzi, że najczęściej kupowane są razem z mlekiem lub sodą, czyli pozostałymi napojami.

{} => {canned beer}

Aż 764 razy, klient, który kupował canned beer, wyszedł ze sklepu tylko z tym produktem.

Rekomendacje dla sklepu:

Celem analizy jest znalezienie produktów, które klienci najczęściej kupują razem. Powyżej przedstawione zostały reguły, które odzwierciedliły wybory klientów w bazie danych. Na tej podstawie powstała mapa sklepu, z rozmieszczeniem produktów, w taki sposób, aby produkty wspólnie kupowane były na przeciwległych końcach, co zmusza konsumenta do obejścia całego sklepu.

  1. Wejście umieszcznone w rogu sklepu sugeruje klientom poruszanie się po strzałkach.
  2. Mleko jest najczęściej kupowanym produktem, dlatego musi znaleść się na końcu sklepu.
  3. Jako, że klienci kupują mleko razem z owocami i warzywami, umieszczono je po dwóch stronach sklepu, dokładnie tak samo jak pieczywo i kiełbaski.
  4. Klienci, kupując piwo, zazwyczaj nie decydowali się na nic więcej, dlatego obok półki z alkoholem rozłozono wyspy z przekąskami, co zwiększy szanse na ich wybór.
  5. warzywa i owoce rozłożone są jako duża wyspa, tak aby klient miał dostęp do wszystkich produktów w jednym miejscu, natomias musi okrążyć stoisko do okoła, co zwiększy szanse na zdecydowanie się na zakup kolejnych produktów.