A comprehensive study of major techniques of multi level frequent pattern mining a survey

IJRET: International Journal of Research in Engineering and Technology ISSN: 2319-1163
__________________________________________________________________________________________
Volume: 02 Issue: 04 | Apr-2013, Available @ http://www.ijret.org 511
A COMPREHENSIVE STUDY OF MAJOR TECHNIQUES OF MULTI
LEVEL FREQUENT PATTERN MINING: A SURVEY
Syed Zishan Ali1
, Yogesh Rathore2
1
Computer Science and Engineering, 2
Professor, Computer Science and Engineering, 1, 2
Raipur Institute of Technology,
Raipur, Mandir Hasaud, Raipur, Chhattisgarh, INDIA
Abstract
Frequent pattern mining has become one of the most popular data mining approaches for the analysis of purchasing patterns. There
are techniques such as Apriority and FP-Growth, which were typically restricted to a single concept level. We extend our research to
study Multi - level frequent patterns in multi-level environments. Mining Multi-level frequent pattern may lead to the discovery of
mining patterns at different levels of hierarchy. In this study, we describe the main techniques used to solve these problems and give
a comprehensive survey of the most influential algorithms
That were proposed during the last decade.
Index Terms: Data Mining, Data Transformation, Frequent Pattern Mining (FPM), Transactional Database.
----------------------------------------------------------------------***------------------------------------------------------------------------
1. INTRODUCTION
Frequent patterns are itemsets, subsequences, or substructures
that appear in a data set with frequency no less than a user-
specified threshold. For example, a set of items, such as milk
and bread, that appear frequently together in a Transaction
data set, is a frequent itemset. Finding frequent patterns plays
an essential role in mining associations, correlations, and
many other interesting relationships among data. Moreover, it
helps in data indexing, classification, clustering, and other data
mining tasks as well. Thus, frequent pattern mining has
become an important data mining task and a focused theme in
data mining research.
Frequent pattern mining was first proposed by Agrawal [1]
for market basket analysis in the form of association rule
mining. It analyses customer buying habits by finding
associations between the different items that customers place
in their “shopping baskets”. For instance, if customers are
buying milk, how likely are they going to also buy cereal (and
what kind of cereal) on the same trip to the supermarket? Such
information can lead to increased sales by helping retailers do
selective marketing and arrange their shelf space
One approach to multilevel mining would be to directly
exploit the standard algorithms in this area – Apriori [1] and
FP-Growth [2] by iteratively applying them in a level by level
manner to each concept level. In this paper, we focused on
the study of frequent patterns based on the FP- tree [3].
Many scholars have published a tons of research work on
frequent pattern mining. There have been extensive studies on
the improvements or extensions of Pattern mining has been
extensively studied in data mining communities for many
years. A variety of efficient algorithms such as PrefixSpan
[4],[5], FP-tree [6],[7] have been proposed. These research
works have mainly focused on developing efficient mining
algorithms for discovering patterns from a large data
collection. However, searching for useful and interesting
patterns and rules was still an open problem [8]. Some of the
basic mining techniques : Apriori, Fp-Growth etc.
2. APRIORI ALGORITHM
Apriori is a algorithm proposed by R. Agrawal and R Srikant
in 1994 [1] for mining frequent item sets for Boolean
association rule. The name of algorithm is based on the fact
that the algorithm uses prior knowledge of frequent item set
properties, as we shall see following. Apriori employs an
iterative approach known as level-wise search, where k item
set are used to explore (k+1) item sets. There are two steps in
each iteration. The first step generates a set of candidate item
sets. Then, in the second step we count the occurrence of
each candidate set in database and prunes all disqualified
candidates (i.e. all infrequent item sets). Apriori uses two
pruning technique, first on the bases of support count (should
be greater than user specified support threshold) and second
for an item set to be frequent , all its subset should be in last
frequent item set The iterations begin with size 2 item sets
and the size is incremented after each iteration. The algorithm
is based on the closure property of frequent item sets: if a set
of items is frequent, then all its proper subsets are also
frequent.
3. FP-GROWTH ALGORITHM
FP-growth [9] is a well-known algorithm that uses the FP- tree
data structure to achieve a condensed representation of the

__________________________________________________________________________________________
database transactions and employs a divide and-conquer
approach to decompose the mining problem into a set of
smaller problems. In essence, it mines all the frequent itemsets
by recursively finding all frequent itemsets in the conditional
pattern base which is efficiently constructed with the help of a
node link structure. A variant of FP-growth is the H-mine
algorithm [10]. It uses array-based and trie-based data
structures to deal with sparse and dense datasets respectively.
FPgrowth* [11] uses an array technique to reduce the FP-tree
traversal time. In FP-growth based algorithms, recursive
construction of the FP-tree affects the algorithm’s
performance.
TID Items bought (ordered) frequent items
100 {z, s, i, d, g, r, e, n} {z , i, s, e, n}
200 {s, h, i, z, l, e, o} {z , i, s, h, e}
300 {h, z, t, j, o, w} {z , h}
400 {h, i , k, s, n} {i, h, n}
500 {s, z, i , e, l ,u, e, n} {z , i, s, e, n}
1. Scan DB once, find frequent 1-itemset (single item
pattern)
2. Sort frequent items in frequency descending order, f-
list
3. Scan DB again, construct FP-tree
TABLE 1:
min_support = 3
F-list= z-i-s-h-e-n
4. METHODOLOGY USED
A. ASCENDING FREQUENCY ORDERED PREFIX-TREE
(AFOPT)
AFOPT is an efficient algorithm for mining frequent itemsets.
It adopts the pattern growth approach and uses a compact data
structure---Ascending Frequency Ordered Prefix-Tree
(AFOPT) to represent the conditional databases[12]. The
AFOPT tree structure is traversed top-down. Compared with
the descending frequent order and bottom-up traversal strategy
adopted by the FP-growth algorithm, the combination of the
top-down traversal strategy and ascending frequency ordering
method requires less pointers to be maintained at each node
and it also reduces the traversal cost of individual conditional
databases.
The goal of mining frequent closed itemsets or maximal
frequent itemsets is to reduce output size. An itemset is closed
if all of its supersets are less frequent than it. An itemset is
maximal if none of its superset is frequent. The complete set
of frequent itemsets can be recovered from the set of frequent
closed itemsets or the set of maximal frequent itemsets. From
frequent closed itemsets, the support information of itemsets
can be recovered, but it cannot be recovered from maximal
frequent itemsets.
B. ADAPTIVE AFOPT ALGORITHM - ADA AFOPT
An algorithm, named ADA AFOPT, which can be used to
resolve the multilevel frequent pattern mining problem. The
algorithm is obtained by extending the AFOPT algorithm for
multi-level databases. The features of the AFOPT
algorithm[12] i.e. FP-Tree, FP-Tree based pattern fragment,
1item partition based divide and conquer method are well
preserved. This algorithm uses flexible support constraints. To
avoid the problem caused by uniform support threshold,
mining with various support constraints is used. Uniform
support threshold might cause problems of either generating
uninteresting patterns at higher abstraction level or missing
potential interesting patterns at lower abstraction are level. At
each level, we classify individual items into two categories:
normal items and exceptional items.
ADA AFOPT algorithm favours users by pushing these
various n transactions, support constraints deep into the
mining process. The interestingness of the patterns mound
generated, hence, is improved dramatically. Being based on
the AFOPT algorithm, this algorithm first traverses the
original database to find frequent items or abstract levels and
sorts them in ascending frequency order. Then the original
database is scanned the second time to construct an AFOPT
structure to represent the conditional databases of the frequent
items.
Header Table
Item frequency
head
z
4
i
4
s
3
h
3
{}
z:4 i:1
i:3 h:1 h:1
s:3 n:3
e:2 h:1
n:2 e:1

__________________________________________________________________________________________
C. TRANSACTION REDUCTION TECHNIQUE
Theorem: If c € Fk and c.support < min.support, Titems ≤ k, k
= 1, then c is useless in Fk+1 where Fk is Frequent pattern, c
is an itemset in each transaction and Titems is total item count
in each transactions.
Proof 1: [For c € Fk ]
Consider a transaction Ti = {T1, T2, T3… Tm}. Let T1 = {a1,
a2, a3… an}, T2 = {a1}. Since c is a hierarchy data.
Whenever the lower-level items of c achieve a support count,
the higher –level items should be added into the reduced
transaction table. For Example data c is 111, 211... Satisfies
support count, therefore the higher –level items of 11*, 21*…
and the next higher-level also 1**, 2** should be added into
the reduced transaction table. If lower-level items of c
does not satisfy the min.support, then lower-level items
of c is removed from the reduced transaction table. Hence
the proof.
Proof 2: [For T items ≤ k where k = 1]
Now consider the same transaction Ti = {T1, T2, T3… Tm}.
Let T1 = {a1, a2, a3… an}, T2 = {a1}. During frequent k+1
pattern generation transaction T2 requires at least 2 as
item count and if not then, Ti can be rejected from the
transaction table. Hence the proof.
Let us consider the following Example with sample database.
TABLE: 1 Sample Database
CCB – Tree Algorithm [13] has been used to find multilevel
frequent 1 pattern.
1:7:7 2:5:7 3:3:8 4:3:8 5:2:8 7:1
11:6:6 12:4:5 13:1:7 21:3:5 22:4:5 32:3:8 41:3:8
52:2:8 71:1
121:2 122:2 131:1 211:3 221:3 222:1 323:3 411:2 413:1 524:2
713:1
111:4 112:1 113:1
After generating the FP tree the next step is to generate
candidate itemsets and find frequent patterns. It begins by
scanning the tree and identifying its leaf nodes. A pointer
to each leaf is then inserting into the leaf node array.
After that a bottom up scan of each leaf node is done
until it reaches the root. Meanwhile each node visited is
conserved into temporary buffer for recording the passing
path when a node with support count is visited. Candidate
Generation keeps the path from starting node i.e. leaf node
to the current node and generate all combinations of
candidate 2-itemset. Only items from all levels that are
above this threshold can be considered as frequent.
Candidate itemset which satisfies the minimum support count
that candidate can be used for next level processing, the node
which does not satisfy minimum .support can be ignored and
candidate generation does nothing for this. After finding
frequent 2-itemsets from all sub trees. Next traversal is
Candidate generation for frequent 3 itemsets. The supports for
all the candidate k-itemsets (k≥3) can be computed and
the frequent k-itemset can be obtained. This process
proceeds until to find frequent k patterns.
TID Items
T1 {111,121,211,221}
T2 {111,211,222,323}
T3 {112,122,221,411}
T4 {111,121}
T5 {111,122,211,221,413}
T6 {113,323,524}
T7 {131,231}
T8 {323,411,524,713}
Root

__________________________________________________________________________________________
TABLE: 2
5. COMPARISON
Table 2 shows the comparison between the techniques that are
discussed in this paper. The characteristics used to distinguish
are: pre-processing, feature extraction, database and result.
Pre-processing is the main step in frequent pattern mining..
Second characteristic is feature extraction which gives the
extracted features that are used in classification. In AFOPT
and ADA-AFOPT the tree is sorted and arranged in ascending
order. Transaction Reduction Technique based method is
used to reduce the unwanted candidates and transactions and
applying the resulted transactions in FP-tree as input to
subsequent iterations of the mining process. It reduces the
I/O costs and search spaces without losing any patterns.
CONCLUSIONS
This paper’s objective is to present the major techniques of
multilevel frequent pattern mining. This paper surveys some
of the important techniques. The techniques considered in this
paper are Ascending Frequency Ordered Prefix-Tree
(AFOPT), Adaptive Afopt algorithm (ADA AFOPT),
Transaction Reduction Technique. The experimental results
shows that AFOPT with ADA AFOPT gives excellent result,
Transaction Reduction Technique are also good in minimizing
I/O cost.
REFERENCES
[1] Agrawal R, Imielinski T, Swami A (1993) Mining
association rules between sets of items in large
databases.In:Proceedings of the1993 ACM-SIGMOD
international conference on management of data
(SIGMOD’93), Washington, DC, pp 207–216,.
[2] Han .J ,Pei .J, and Yin .Y,(2000) Mining Frequent
patterns without candidate generation. In Proc. Of
ACM- SIGMOD Int. Conf. on Management of
Data, pages 1- 12.
[3] T.Eavis and XI Zheng, Multi-Level Frequent
Pattern Mining, in Springer-Verlag Berlin Heidelberg
2009, pp. 369 – 383..
[4] J. Pei, J. Han, B. Mortazavi-Asl, H. Pinto, Q. Chen, U.
Dayal, and M. Hsu, “Prefixspan: Mining Sequential
Patterns Efficiently by Prefix-Projected Pattern
Growth,” Proc. 17th Int’l Conf. Data Eng. (ICDE ’01),
pp. 215-224, 2001.
[5] X. Yan, J. Han, and R. Afshar, “Clospan: Mining
Closed Sequential Patterns in Large Datasets,” Proc.
SIAM Int’l Conf. Data Mining (SDM ’03), pp. 166-
177, 2003
[6] J. Han and K.C.-C. Chang, “Data Mining for Web
Intelligence,” Computer, vol. 35, no. 11, pp. 64-70,
Nov. 2002.
[7] J. Han, J. Pei, and Y. Yin, “Mining Frequent Patterns
without Candidate Generation,” Proc. ACM SIGMOD
Int’l Conf. Management of Data (SIGMOD ’00), pp. 1-
12, 2000.
[8] Y. Li and N. Zhong, “Interpretations of Association
Rules by Granular Computing,” Proc. IEEE Third Int’l
Conf. Data Mining (ICDM ’03), pp. 593-596, 2003.
[9] J. Han, J. Pei, and Y. Yin,“Mining Frequent Patterns
without Candidate Generation,” Proceedings of ACM
SIGMOD International Conference on Management of
Data, ACM Press, Dallas, Texas, pp. 1-12, May 2000.
[10] J. Pei, J. Han, H. Lu, S. Nishio, S. Tang, and D. Yang,
“Hmine: Hyper Structure Mining of Frequent Patterns
in Large Databases,” Proceedings of IEEE
International Conference on Data Mining, pp. 441-448,
2001.
[11] G. Grahne, and J. Zhu, ”Efficiently using prefix-trees in
mining frequent itemsets,” FIMI ’03, Frequent Itemset
Mining Implementations, Proceedings of the ICDM
2003 Workshop on Frequent Itemset Mining
Implementations, Melbourne, Florida,December 2003.
[12] Int. J. of Computers, Communications & Control, ISSN
1841-9836, E-ISSN 1841-9844 Vol. III (2008), Suppl.
issue: Proceedings of ICCCC 2008, pp. 437-441
[13] Dr.K.Duraiswamy and B.Jayanthi, a Novel
preprocessing Algorithm for Frequent Pattern Mining
in Mutidatasets, International Journal of Data
Engineering,Vol. 2, No. 3, Aug 2011.
Techniques
Pre-
processing
Feature
extraction
Result
Ascending
Frequency
Ordered Prefix-
Tree (AFOPT)
To Traverse
the trees in
top-down
depth-ﬁrst
order
Sorted
tree(Ascendin
g order)
Feasible and
effective
ADAPTIVE
AFOPT
algorithm - ADA
AFOPT
To Scan the
traversed tree
Scanned Tree
Excellent
Transaction
Reduction
Technique
To Reduce
Non candidate
Pattern
Reduced
Search tree
with less I/O
cost.
Good

__________________________________________________________________________________________
BIOGRAPHIES
Syed Zishan Ali is with the Department of
Computer Science and Engineering, hilai
Institute of Technology, Raipur, Chhattisgarh,
India.. E-mail: zishan786s@gmail.com
Yogesh Rathore is currently the coordinator
of M.Tech, and is a Senior lecturer in
Department of Computer Science and
engineering, Raipur Institute of Technology.
Raipur, Chhattisgarh, India. E-mail:
yogeshrathore23@gmail.com

A comprehensive study of major techniques of multi level frequent pattern mining a survey

More Related Content

What's hot (20)

Viewers also liked (16)

Similar to A comprehensive study of major techniques of multi level frequent pattern mining a survey (20)

More from eSAT Journals (20)

Recently uploaded (20)

A comprehensive study of major techniques of multi level frequent pattern mining a survey