Apriori Sets And Sequences - Keith's MS Thesis
Apriori Sets And Sequences
About
Code
Performance Data Collection
·
Data Sets
Documents
Results
·
                                          
Printer Friendly Version
Results - Rules and Performance

Intro ] 2002-06-26-even-better-rules ] 2002-07-27-66 ]
2002-07-29-99 ] 2003-03-03-random-test ] Test-Matrix.xls ]
compare-duplicate-lookup-vector-tree-1 ] [ compare-previous-dups-no-previous-1 ] purpose ]
reduced-k=1-2002-05-09 ] trace-2002-05-03-16:53 ] very first rules ever ]

caches ] [ old-incomplete ] [ performance-improvements ] [ promoters ] [ sleep ] [ stock-market ]

--- No previous duplicate hash ---
Beginning to mine with min support 0.95...
Level 1 candidates: 169 (117)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 2930 comparisons.
21 seconds: # of frequent itemsets for level 1 = 21
`(0/0)
ARMinerApriori.generateCandidates: pruned 0 itemsets using checkSubsets.
Possible sets of duplicates: 0.
3 seconds: # of Generated candidates for level: 2 = 2046 (2046)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 55818 comparisons.
426 seconds: # of frequent itemsets for level 2 = 82
`(0/0)
ARMinerApriori.generateCandidates: pruned 5972 itemsets using checkSubsets.
Possible sets of duplicates: 0.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 2507
7 seconds: # of Generated candidates for level: 3 = 142 (142)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 0 comparisons.
61 seconds: # of frequent itemsets for level 3 = 142
`(0/0)`(117/117)
ARMinerApriori.generateCandidates: pruned 6144 itemsets using checkSubsets.
Possible sets of duplicates: 0.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 8568
13 seconds: # of Generated candidates for level: 4 = 133 (133)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 0 comparisons.
71 seconds: # of frequent itemsets for level 4 = 133
`(0/0)`(65/65)
ARMinerApriori.generateCandidates: pruned 2244 itemsets using checkSubsets.
Possible sets of duplicates: 0.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 7521
8 seconds: # of Generated candidates for level: 5 = 73 (73)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 0 comparisons.
51 seconds: # of frequent itemsets for level 5 = 73
`(0/0)
ARMinerApriori.generateCandidates: pruned 128 itemsets using checkSubsets.
Possible sets of duplicates: 0.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 2112
2 seconds: # of Generated candidates for level: 6 = 24 (24)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 0 comparisons.
27 seconds: # of frequent itemsets for level 6 = 24
`(0/0)
ARMinerApriori.generateCandidates: pruned 24 itemsets using checkSubsets.
Possible sets of duplicates: 0.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 192
2 seconds: # of Generated candidates for level: 7 = 4 (4)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weightCandidates: removing candidates saved 0 comparisons.
16 seconds: # of frequent itemsets for level 7 = 4
`(0/0)
ARMinerApriori.generateCandidates: pruned 4 itemsets using checkSubsets.
Possible sets of duplicates: 0.
2 seconds: # of Generated candidates for level: 8 = 0 (0)
AprioriRules.findAssociations: 1704 rules. 186 duplicate rules removed. 1518 rules left.

Mining Started : 1046367547257
Mining Complete: 1046368326340
time taken: 779 seconds

Required Attributes in Antecedents:
none

Required Attributes in Consequents:
none

--- /No previous duplicate hash ---

--- Yes previous duplicate hash ---

Beginning to mine with min support 0.95...
Level 1 candidates: 169 (117)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 2930 comparisons.
23 seconds: # of frequent itemsets for level 1 = 21
`(0/0)
ARMinerApriori.generateCandidates: pruned 0 itemsets using checkSubsets.
Possible sets of duplicates: 596.
4 seconds: # of Generated candidates for level: 2 = 2046 (722)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 19415 comparisons.
133 seconds: # of frequent itemsets for level 2 = 82
`(0/0)
ARMinerApriori.generateCandidates: pruned 5972 itemsets using checkSubsets.
Possible sets of duplicates: 52.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 2507
8 seconds: # of Generated candidates for level: 3 = 142 (62)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 0 comparisons.
32 seconds: # of frequent itemsets for level 3 = 142
`(0/0)`(117/45)
ARMinerApriori.generateCandidates: pruned 6144 itemsets using checkSubsets.
Possible sets of duplicates: 45.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 8568
13 seconds: # of Generated candidates for level: 4 = 133 (50)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 0 comparisons.
34 seconds: # of frequent itemsets for level 4 = 133
`(0/0)`(65/22)
ARMinerApriori.generateCandidates: pruned 2244 itemsets using checkSubsets.
Possible sets of duplicates: 23.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 7521
6 seconds: # of Generated candidates for level: 5 = 73 (24)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 0 comparisons.
26 seconds: # of frequent itemsets for level 5 = 73
`(0/0)
ARMinerApriori.generateCandidates: pruned 128 itemsets using checkSubsets.
Possible sets of duplicates: 7.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 2112
3 seconds: # of Generated candidates for level: 6 = 24 (7)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 0 comparisons.
17 seconds: # of frequent itemsets for level 6 = 24
`(0/0)
ARMinerApriori.generateCandidates: pruned 24 itemsets using checkSubsets.
Possible sets of duplicates: 1.
ARMinerApriori.generateCandidates: Calls to getCandidate saved: 192
2 seconds: # of Generated candidates for level: 7 = 4 (1)
ARMinerApriori.weighCandidates: counting supports: 33 rows, 1:1 - 
|||||||||||||||||||||||||||||||||
ARMinerApriori.weighCandidates: removing candidates saved 0 comparisons.
14 seconds: # of frequent itemsets for level 7 = 4
`(0/0)
ARMinerApriori.generateCandidates: pruned 4 itemsets using checkSubsets.
Possible sets of duplicates: 0.
2 seconds: # of Generated candidates for level: 8 = 0 (0)
AprioriRules.findAssociations: 1704 rules. 186 duplicate rules removed. 1518 rules left.

Mining Started : 1046373219685
Mining Complete: 1046373609120
time taken: 390 seconds

Required Attributes in Antecedents:
none

Required Attributes in Consequents:
none

--- /Yes previous duplicate hash ---


--- Line by line compare ---

No : 21 seconds: # of frequent itemsets for level 1 = 21
Yes: 23 seconds: # of frequent itemsets for level 1 = 21

No : 3 seconds: # of Generated candidates for level: 2 = 2046 (2046)
Yes: 4 seconds: # of Generated candidates for level: 2 = 2046 (722)

No : 426 seconds: # of frequent itemsets for level 2 = 82
Yes: 133 seconds: # of frequent itemsets for level 2 = 82

No : 7 seconds: # of Generated candidates for level: 3 = 142 (142)
Yes: 8 seconds: # of Generated candidates for level: 3 = 142 (62)

No : 61 seconds: # of frequent itemsets for level 3 = 142
Yes: 32 seconds: # of frequent itemsets for level 3 = 142

No : 13 seconds: # of Generated candidates for level: 4 = 133 (133)
Yes: 13 seconds: # of Generated candidates for level: 4 = 133 (50)

No : 71 seconds: # of frequent itemsets for level 4 = 133
Yes: 34 seconds: # of frequent itemsets for level 4 = 133

No : 8 seconds: # of Generated candidates for level: 5 = 73 (73)
Yes: 6 seconds: # of Generated candidates for level: 5 = 73 (24)

No : 51 seconds: # of frequent itemsets for level 5 = 73
Yes: 26 seconds: # of frequent itemsets for level 5 = 73

No : 2 seconds: # of Generated candidates for level: 6 = 24 (24)
Yes: 3 seconds: # of Generated candidates for level: 6 = 24 (7)

No : 27 seconds: # of frequent itemsets for level 6 = 24
Yes: 17 seconds: # of frequent itemsets for level 6 = 24

No : 2 seconds: # of Generated candidates for level: 7 = 4 (4)
Yes: 2 seconds: # of Generated candidates for level: 7 = 4 (1)

No : 16 seconds: # of frequent itemsets for level 7 = 4
Yes: 14 seconds: # of frequent itemsets for level 7 = 4

No : 2 seconds: # of Generated candidates for level: 8 = 0 (0)
Yes: 2 seconds: # of Generated candidates for level: 8 = 0 (0)

No : time taken: 779 seconds
Yes: time taken: 390 seconds

--- /Line by line compare ---

It seems as though there is little memory overhead to use the
previous duplicate hash but we save time. 

this test was invalid because we were not detecting duplicates using
the prefix tree. Not the mismatching unique candidate itemset counts.

by: Keith A. Pray
Last Modified: July 4, 2004 8:02 AM
© 2004 - 1975 Keith A. Pray.
All rights reserved.

Current Theme: