++ed by:

3 non-PAUSE users.

Alex Graciano

NAME

Data::Mining::Apriori - Perl extension for implement the apriori algorithm of data mining.

SYNOPSIS

        use strict;
        use warnings;
        use Data::Mining::Apriori;

        # TRANSACTION 103:CEREAL 101:MILK 102:BREAD
        #        1101          1        1         0
        #        1102          1        0         1
        #        1103          1        1         1
        #        1104          1        1         1
        #        1105          0        1         1
        #        1106          1        1         1
        #        1107          1        1         1
        #        1108          1        0         1
        #        1109          1        1         1
        #        1110          1        1         1

        my $apriori = new Data::Mining::Apriori;

        $apriori->{metrics}{minSupport}=0.0155; # The minimum support (required), default value is 0.01 (1%)

        $apriori->{metrics}{minConfidence}=0.0155; # The minimum confidence (required), default value is 0.10 (10%)

        $apriori->{metrics}{minLift}=1; # The minimum lift (optional)

        $apriori->{metrics}{minLeverage}=0; # The minimum leverage (optional)

        $apriori->{metrics}{minConviction}=0; # The minimum conviction (optional)

        $apriori->{metrics}{minCoverage}=0; # The minimum coverage (optional)

        $apriori->{metrics}{minCorrelation}=0; # The minimum correlation (optional)

        $apriori->{metrics}{minCosine}=0; # The minimum cosine (optional)

        $apriori->{metrics}{minLaplace}=0; # The minimum laplace (optional)

        $apriori->{metrics}{minJaccard}=0; # The minimum jaccard (optional)

        $apriori->{precision}=2; # Sets the floating point precision of the metrics (required), default value is 3

        $apriori->{output}=1;
        # The output type (optional): 1 - Export to text file delimited by TAB; 2 - Export to excel file with chart.

        $apriori->{pathOutputFiles}='data/'; # The path to output files (optional)

        $apriori->{messages}=1; # A value boolean to display the messages (optional)

        $apriori->{keyItemsDescription}{'101'}='MILK'; # Hash table reference to add items by key and description
        $apriori->{keyItemsDescription}{102}='BREAD';
        $apriori->{keyItemsDescription}{'103'}='CEREAL';

        my@items=(103,101);
        $apriori->insert_key_items_transaction(\@items); # Insert key items by transaction
        $apriori->insert_key_items_transaction([103,102]);
        $apriori->insert_key_items_transaction([103,101,102]);
        $apriori->insert_key_items_transaction([103,101,102]);
        $apriori->insert_key_items_transaction([101,102]);
        $apriori->insert_key_items_transaction([103,101,102]);
        $apriori->insert_key_items_transaction([103,101,102]);
        $apriori->insert_key_items_transaction([103,102]);
        $apriori->insert_key_items_transaction([103,101,102]);
        $apriori->insert_key_items_transaction([103,101,102]);

        # or from a data file

        $apriori->input_data_file("datafile.txt",",");
        # Insert key items by line (transaction), accepts the arguments of path to data file and item separator

        # file contents (example)

        103,101
        103,102
        103,101,102
        103,101,102
        101,102
        103,101,102
        103,101,102
        103,102
        103,101,102
        103,101,102

        print "\n${\$apriori->quantity_possible_rules}"; # Show the quantity of possible rules

        $apriori->{limitRules}=10; # The limit of rules (optional)

        $apriori->{limitSubsets}=12; # The limit of subsets (optional)

        $apriori->generate_rules;
        # Generate association rules to no longer meet the minimum support, confidence, lift, leverage, conviction, coverage, correlation, cosine, laplace, jaccard or limit of rules

        print "\n@{$apriori->{frequentItemset}}\n"; # Show frequent items

        #output messages

        12
        3 items, 12 possible rules
        Large itemset of length 2, 3 items
        Processing ...
        Frequent itemset: { 102, 103, 101 }, 3 items
        Exporting to file data/output_large_itemset_length_2.txt ...
        Large itemset of length 3, 3 items
        Processing ...
        Frequent itemset: { 101, 102, 103 }, 3 items
        Exporting to file data/output_large_itemset_length_3.txt ...
        101, 102, 103

        #output file "output_itemset_length_2.txt"

        Rules   Support Confidence      Lift    Leverage        Conviction      Coverage        Correlation     Cosine  Laplace Jaccard
        R1      0,80    0,89    1,11    0,08    1,80    0,90    0,67    0,94    0,62    0,89
        R2      0,70    0,78    1,11    0,07    1,35    0,90    0,51    0,88    0,59    0,78
        R3      0,80    0,89    1,11    0,08    1,80    0,90    0,67    0,94    0,62    0,89
        R4      0,70    0,78    1,11    0,07    1,35    0,90    0,51    0,88    0,59    0,78
        R5      0,70    0,87    1,25    0,14    2,40    0,80    0,76    0,94    0,61    0,87
        R6      0,70    0,87    1,25    0,14    2,40    0,80    0,76    0,94    0,61    0,87

        Rule R1: { 102 } => { 103 }
        Support: 0,80
        Confidence: 0,89
        Lift: 1,11
        Leverage: 0,08
        Conviction: 1,80
        Coverage: 0,90
        Correlation: 0,67
        Cosine: 0,94
        Laplace: 0,62
        Jaccard: 0,89
        Items:
        102 BREAD
        103 CEREAL

        #...

        #output file "output_itemset_length_3.txt"

        Rules   Support Confidence      Lift    Leverage        Conviction      Coverage        Correlation     Cosine  Laplace Jaccard
        R7      0,60    0,67    1,11    0,06    1,20    0,90    0,41    0,82    0,55    0,67
        R8      0,60    0,75    1,25    0,12    1,60    0,80    0,61    0,87    0,57    0,75
        R9      0,60    0,86    1,43    0,18    2,80    0,70    0,80    0,93    0,59    0,86
        R10     0,60    0,67    1,11    0,06    1,20    0,90    0,41    0,82    0,55    0,67
        R11     0,60    0,86    1,43    0,18    2,80    0,70    0,80    0,93    0,59    0,86
        R12     0,60    0,75    1,25    0,12    1,60    0,80    0,61    0,87    0,57    0,75

        Rule R7: { 102 } => { 101, 103 }
        Support: 0,60
        Confidence: 0,67
        Lift: 1,11
        Leverage: 0,06
        Conviction: 1,20
        Coverage: 0,90
        Correlation: 0,41
        Cosine: 0,82
        Laplace: 0,55
        Jaccard: 0,67
        Items:
        102 BREAD
        101 MILK
        103 CEREAL

        Rule R8: { 102, 103 } => { 101 }
        Support: 0,60
        Confidence: 0,75
        Lift: 1,25
        Leverage: 0,12
        Conviction: 1,60
        Coverage: 0,80
        Correlation: 0,61
        Cosine: 0,87
        Laplace: 0,57
        Jaccard: 0,75
        Items:
        102 BREAD
        103 CEREAL
        101 MILK

        #...

DESCRIPTION

This module implements the apriori algorithm of data mining.

ATTRIBUTES

totalTransactions

The total number of transactions.

metrics

The type of metrics

minSupport

The minimum support (required), default value is 0.01 (1%)

minConfidence

The minimum confidence (required), default value is 0.10 (10%)

minLift

The minimum lift (optional)

minLeverage

The minimum leverage (optional)

minConviction

The minimum conviction (optional)

minCoverage

The minimum coverage (optional)

minCorrelation

The minimum correlation (optional)

minCosine

The minimum cosine (optional)

minLaplace

The minimum laplace (optional)

minJaccard

The minimum jaccard (optional)

precision

Sets the floating point precision of the metrics (required), default value is 3

limitRules

The limit of rules (optional)

limitSubsets

The limit of subsets (optional)

output

The output type (optional):

  • 1 - Text file delimited by TAB;

  • 2 - Excel file with chart.

pathOutputFiles

The path to output files (optional)

messages

A value boolean to display the messages (optional)

keyItemsDescription

Hash table reference to add item by key and description.

keyItemsTransactions

Hash table reference to add items by keys and transactions.

frequentItemset

Frequent itemset.

associationRules

A data structure to store the name of the rule, key items, implication, support, confidence, lift, leverage, conviction, coverage, correlation, cosine, laplace and jaccard.

        $self->{associationRules} = {
                                                                  '1' => {
                                                                                   'confidence' => '0.89',
                                                                                   'cosine' => '0.94',
                                                                                   'implication' => '{ 102 } => { 103 }',
                                                                                   'coverage' => '0.90',
                                                                                   'laplace' => '0.62',
                                                                                   'jaccard' => '0.89',
                                                                                   'support' => '0.80',
                                                                                   'correlation' => '0.67',
                                                                                   'items' => [
                                                                                                                '102',
                                                                                                                '103'
                                                                                                          ],
                                                                                   'conviction' => '1.80',
                                                                                   'lift' => '1.11',
                                                                                   'leverage' => '0.08'
                                                                                 },
                                                                        #...

METHODS

new

Creates a new instance of Data::Mining::Apriori.

insert_key_items_transaction(\@items)

Insert key items per transaction. Accepts the following arguments:

  • An array reference to key items.

input_data_file("datafile.txt",",")

Insert items per line (transaction). Accepts the following arguments:

  • Data file;

  • Item separator.

        # file contents (example)

        103,101
        103,102
        103,101,102
        103,101,102
        101,102
        103,101,102
        103,101,102
        103,102
        103,101,102
        103,101,102

quantity_possible_rules

Returns the quantity of possible rules.

generate_rules

Generate association rules until no set of items meets the minimum support, confidence, lift, leverage, conviction, coverage, correlation, cosine, laplace, jaccard or limit of rules.

association_rules

Generate association rules by length of large itemsets.

AUTHOR

Alex Graciano, <agraciano@cpan.org>

COPYRIGHT AND LICENSE

Copyright (C) 2015-2018 by Alex Graciano

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.12.4 or, at your option, any later version of Perl 5 you may have available.