Apriori

# Data Preprocessing

# install.packages('arules')

library(arules)

dataset = read.csv('Market_Basket_Optimisation.csv', header = FALSE)

#sparce matrix

dataset = read.transactions('Market_Basket_Optimisation.csv', sep = ',', rm.duplicates = TRUE)

distribution of transactions with duplicates:

1 

5 

summary(dataset)

transactions as itemMatrix in sparse format with

 7501 rows (elements/itemsets/transactions) and

 119 columns (items) and a density of 0.03288973 

most frequent items:

mineral water          eggs     spaghetti  french fries     chocolate       (Other) 

         1788          1348          1306          1282          1229         22405 

element (itemset/transaction) length distribution:

sizes

   1    2    3    4    5    6    7    8    9   10   11   12   13   14   15   16   18   19   20 

1754 1358 1044  816  667  493  391  324  259  139  102   67   40   22   17    4    1    2    1 

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 

  1.000   2.000   3.000   3.914   5.000  20.000 

includes extended item information - examples:

             labels

1           almonds

2 antioxydant juice

3         asparagus

itemFrequencyPlot(dataset, topN = 20)

itemFrequencyPlot(dataset, topN = 10)

# Training Apriori on the dataset

#higher support(hight frequent product) and confidence

rules = apriori(data = dataset, parameter = list(support = 0.003, confidence = 0.8))

output

Apriori

Parameter specification:

 confidence minval smax arem  aval originalSupport maxtime support minlen maxlen

        0.8    0.1    1 none FALSE            TRUE       5   0.003      1     10

 target   ext

  rules FALSE

Algorithmic control:

 filter tree heap memopt load sort verbose

    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 22 

set item appearances ...[0 item(s)] done [0.00s].

set transactions ...[119 item(s), 7501 transaction(s)] done [0.00s].

sorting and recoding items ... [115 item(s)] done [0.00s].

creating transaction tree ... done [0.00s].

checking subsets of size 1 2 3 4 5 done [0.01s].

writing ... [0 rule(s)] done [0.00s].

creating S4 object  ... done [0.00s].

# Visualising the results

inspect(sort(rules, by = 'lift')[1:10])

Error in slot(x, s)[i] : subscript out of bounds

rules = apriori(data = dataset, parameter = list(support = 0.003, confidence = 0.4))

Apriori

Parameter specification:

 confidence minval smax arem  aval originalSupport maxtime support minlen maxlen

        0.4    0.1    1 none FALSE            TRUE       5   0.003      1     10

 target   ext

  rules FALSE

Algorithmic control:

 filter tree heap memopt load sort verbose

    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 22 

set item appearances ...[0 item(s)] done [0.00s].

set transactions ...[119 item(s), 7501 transaction(s)] done [0.00s].

sorting and recoding items ... [115 item(s)] done [0.00s].

creating transaction tree ... done [0.00s].

checking subsets of size 1 2 3 4 5 done [0.01s].

writing ... [281 rule(s)] done [0.00s].

creating S4 object  ... done [0.00s].

lhs                                            rhs                 support    

[1]  {mineral water,whole wheat pasta}           => {olive oil}         0.003866151

[2]  {spaghetti,tomato sauce}                    => {ground beef}       0.003066258

[3]  {french fries,herb & pepper}                => {ground beef}       0.003199573

[4]  {cereals,spaghetti}                         => {ground beef}       0.003066258

[5]  {frozen vegetables,mineral water,soup}      => {milk}              0.003066258

[6]  {chocolate,herb & pepper}                   => {ground beef}       0.003999467

[7]  {chocolate,mineral water,shrimp}            => {frozen vegetables} 0.003199573

[8]  {frozen vegetables,mineral water,olive oil} => {milk}              0.003332889

[9]  {cereals,ground beef}                       => {spaghetti}         0.003066258

[10] {frozen vegetables,soup}                    => {milk}              0.003999467

     confidence lift     count

[1]  0.4027778  6.115863 29   

[2]  0.4893617  4.980600 23   

[3]  0.4615385  4.697422 24   

[4]  0.4600000  4.681764 23   

[5]  0.6052632  4.670863 23   

[6]  0.4411765  4.490183 30   

[7]  0.4210526  4.417225 24   

[8]  0.5102041  3.937285 25   

[9]  0.6764706  3.885303 23   

[10] 0.5000000  3.858539 30   

rules = apriori(data = dataset, parameter = list(support = 0.004, confidence = 0.2))

Apriori

Parameter specification:

 confidence minval smax arem  aval originalSupport maxtime support minlen maxlen

        0.2    0.1    1 none FALSE            TRUE       5   0.004      1     10

 target   ext

  rules FALSE

Algorithmic control:

 filter tree heap memopt load sort verbose

    0.1 TRUE TRUE  FALSE TRUE    2    TRUE

Absolute minimum support count: 30 

set item appearances ...[0 item(s)] done [0.00s].

set transactions ...[119 item(s), 7501 transaction(s)] done [0.01s].

sorting and recoding items ... [114 item(s)] done [0.00s].

creating transaction tree ... done [0.00s].

checking subsets of size 1 2 3 4 done [0.01s].

writing ... [811 rule(s)] done [0.00s].

creating S4 object  ... done [0.00s]. 

# Visualising the results

inspect(sort(rules, by = 'lift')[1:10])

lhs                       rhs                 support confidence     lift count

[1]  {light cream}          => {chicken}       0.004532729  0.2905983 4.843951    34

[2]  {pasta}                => {escalope}      0.005865885  0.3728814 4.700812    44

[3]  {pasta}                => {shrimp}        0.005065991  0.3220339 4.506672    38

[4]  {eggs,                                                                         

      ground beef}          => {herb & pepper} 0.004132782  0.2066667 4.178455    31

[5]  {whole wheat pasta}    => {olive oil}     0.007998933  0.2714932 4.122410    60

[6]  {herb & pepper,                                                                

      spaghetti}            => {ground beef}   0.006399147  0.3934426 4.004360    48

[7]  {herb & pepper,                                                                

      mineral water}        => {ground beef}   0.006665778  0.3906250 3.975683    50

[8]  {tomato sauce}         => {ground beef}   0.005332622  0.3773585 3.840659    40

[9]  {mushroom cream sauce} => {escalope}      0.005732569  0.3006993 3.790833    43

[10] {frozen vegetables,                                                            

      mineral water,                                                                

      spaghetti}            => {ground beef}   0.004399413  0.3666667 3.731841    33