Hu3414421448

R. Sridevi, Dr. E. Ramaraj / International Journal of Engineering Research and Applications
(IJERA) ISSN: 2248-9622 www.ijera.com
Vol. 3, Issue 4, Jul-Aug 2013, pp.1442-1448
1442 | P a g e
A General Survey on Multidimensional And Quantitative
Association Rule Mining Algorithms
R. Sridevi1
, Dr. E. Ramaraj2
1
Asst. Professor, Dept. of Information Technology Dr.Umayal Ramanathan College for Women, Karaikudi
2
Professor, Dept. of Computer Science & Engineering, Alagappa University, Karaikudi.
Abstract
Data mining is one of the significant
topics of research in recent years. Association
rule is a method for discovering interesting
relations between variables in large databases.
Support and Confidence are the two basic
parameters used to study the threshold values
for each database. In this paper, an overall
survey of the algorithms implementing the
multidimensional and quantitative data is
presented. It also illustrates the various
approaches in association rule. The concept of
Bit Mask Search Algorithm can be identified
and suggested to effectively generate the
frequent Item sets from large database.
Moreover, the algorithm is best suited for
multidimensional and quantitative datasets. It
describes the essential role of multidimensional
and Quantitative data in Association rule
mining algorithm.
Keywords: Data Mining, Association rules,
Multidimensional rule, Quantitative rule.
I. INTRODUCTION
Data mining is a non trivial extraction of
implicit, previously unknown and potentially useful
information from data. Data mining technology
provides a user- oriented approach to novel and
hidden patterns in the data [3]. Data mining refers
to extracting or “mining” knowledge from large
amounts of data.
Discovery of hidden patterns and
relationships often goes unexploited. Advanced
data mining techniques can help to overcome this
situation. Data mining technology provides a user-
oriented approach to novel and hidden patterns in
the data. A wide variety of areas including
marketing, customer relationship management,
engineering, medicine, crime analysis, expert
prediction, Web mining, and mobile computing,
besides others utilize Data mining.
The essential process of Knowledge
Discovery is the conversion of data into knowledge
in order to aid in decision making, referred to as
data mining. Knowledge Discovery process
consists of an iterative sequence of data cleaning,
data integration, data selection, data mining, pattern
recognition and knowledge presentation.
The difficulties posed by prediction problems have
resulted in a variety of problem-solving techniques.
For example, data mining methods comprise
artificial neural networks and decision trees, and
statistical techniques including linear regression
and stepwise polynomial regression [8]. Data
mining technology provides a user oriented
approach to novel and hidden patterns in the data.
Data Mining have two flavors- directed and
undirected. Directed data mining attempts to
explain or categorize some particular target field
such as income or response. Undirected data
mining attempts to find patterns of similarities
among groups of records without the use of a
particular target field or collection of predefined
classes. Data mining is largely concerned with
building models. A model is simply an algorithm or
set of levels that connects a collection of inputs to a
particular target or outcome.
Data analysis is a process in which raw
data is prepared and structured so that valuable
information can be extracted from it. The process
of organizing and thinking about data is way to
accept what the data does and does not contain.
There are a variety of ways in which public can
approach data analysis. In fact, most data mining
algorithms require large amounts of data in order to
build and train the models that will then be used to
perform classification, prediction, estimation, or
other data mining tasks.
II. DATA MINING TECHNIQUES
Basically, data mining is about processing
data and identifying the patterns in that information
so that it is easy to decide or judge. Data mining
principles have been around for many years, but,
with the advent of large amount of data, it is even
more prevalent. Large data caused an explosion in
the use of more extensive data mining techniques,
partially because the size of the information is
much larger and because the information tends to
be more varied and extensive in its nature and
content. With large data sets, it is no longer enough
to get relatively simple and straightforward
statistics out of the system.
2.1 ASSOCIATION RULES TYPES
Association rule mining, one of the most important
and well researched techniques of data mining, was

1443 | P a g e
first introduced in [2]. Association rules are widely
used in various areas such as telecommunication
networks, market and risk management, inventory
control etc. Association rule mining is to find out
association rules that satisfy the predefined
minimum support and confidence from a given
database. The problem is usually decomposed into
two sub-problems. One is to find those item sets
whose occurrences exceed a predefined threshold
in the database; those item sets are called frequent
or large item sets. The second problem is to
generate association rules from those large item
sets with the constraints of minimal confidence
[4].The various types of association rules are
clearly described in the following section.
Each item can be seen as a Boolean
variable presenting the absence or presence of that
item in the transaction or in the row. Binary
association rule in databases of items can be
organized and clustered [13]. By calculating a
Minimum Spanning Tree on the graph of
associations, the most significant associations can
be discovered and easily visualized, allowing easy
understanding of existing relations.
Quantitative Association rule
In practice, database contains much
quantitative data and is not limited to categorical
items only. Quantitative association rules (QARs)
is an important tool in discovering association
relationships among sets of attributes in business
and scientific domains [13]. In QAR, attributes are
not limited to being Boolean but can either be
quantitative which numeric values are (eg. Age,
salary) or categorical, which are enumerations (eg.
Gender, name).
Fuzzy Association Rule
The new fuzzy association rule mining
approach [18] emerged out of the necessity to mine
quantitative data frequently present in databases
efficiently. When dividing an attribute in the data
into sets covering certain ranges of values, the
users are confronted with the sharp boundary
problem. Elements near the boundaries of a crisp
set will either be ignored or overemphasized.
Multilevel Association Rule
Multilevel association rules are used to
find the frequently occurring patterns and
reasonably strong rule implications. A user or an
expert may specify two thresholds: minimum
support and minimum confidence. The multilevel
association rules can be professionally mined
using concept hierarchies, which defines a series of
mappings from a set of low level concepts to
higher-level with more general concepts[12]. On
lowest levels, it strength will be that no rules may
match the constraints. At highest levels, rules can
be tremendously universal. Usually, a top down
proceed is used where the support threshold varies
from level to level.
Hegland [7] reviews the most well known
algorithm for producing association rules such as
Apriori and discuss variants for distributed data,
inclusion of constraints and data taxonomies. The
review ends with an outlook on tools which have
the potential to deal with long item sets and
considerably reduce the amount of (uninteresting)
item sets returned.
2.2 ASSOCIATION RULE MINING
Boolean Association Rule
In this section association rule mining
problem is discussed in detail. Different issues in
Association Rule Mining (ARM) will be elaborated
together with classic algorithms which are
presented in this paper.
Basic Concepts & Basic Association Rules
Algorithms
The formal statement of association rule
mining problem was stated by Agrawal in [1]. Let
I=I1, I2, … , Im be a set of m distinct attributes, T
be transaction that contains a set of items such that
T ⊆ I, D be a database with different transaction
records Ts. An association rule is an implication in
the form of X⇒Y, where X, Y ⊂ I are sets of items
called item sets, the rule means X implies Y. There
are two important basic measures for association
rules, support(s) and confidence(c).
Support(s) of an association rule is defined
as the percentage/fraction of records that
contains X and Y (XUY) to the total number of
records in the database. The count for each
item is increased by one every time the item
is encountered in different transaction T in
database D during the scanning process. It
means the support count does not take the
quantity of the item into account.
Support (X⇒Y) = P (XUY)
N
Confidence (X⇒Y) = P (XUY)
P(X)
Where X, Y is the measure of Antecedent and
Consequent, P is the Probability at which it can
occur and N is the total number of records.
Apriori Algorithm

1444 | P a g e
Apriori is a great improvement in the history of
association rule mining, which was first proposed
by Agrawal in [Agrawal and Srikant
1994].Initially, the candidate itemsets are
generated, then the database is scanned to check the
actual support count of the corresponding itemsets.
During the first scanning of the database the
support count of each item is calculated and the
large 1 -itemsets are generated by pruning those
itemsets whose supports are below the pre-
defined threshold values. The candidates are
generated by joining among the frequent itemsets
level-wise, also candidate are pruned according the
Apriori property [14].
However there are two bottlenecks in the Apriori
algorithm. One is the complex candidate generation
process that uses most of the time, space and
memory. Another one is the multiple scan of the
database. Based on Apriori algorithm, many new
algorithms were designed with some modifications
or improvements.
FP-Tree(Frequent Pattern Tree) Algorithm
FP-Tree [14], frequent pattern mining, is
another milestone in the development of
association rule mining, which breaks the two
bottlenecks of the Apriori. The frequent itemsets
are generated with only two passes over the
database and without any candidate generation
process. The process of constructing the FP-Tree is
as follows.
Step I: The database is scanned for the first time,
during this scanning the support count of each
items are collected. As a result the frequent 1-
itemsets of this process is the same as in Apriori
algorithm. Those frequent itemsets are sorted in a
descending order with their supports. Also the head
table of ordered frequent 1 -itemsets is created.
Step II: Create the root node of the FP-Tree T with
a label of Root. The database is scanned again to
construct the FP-Tree with the head table, for each
transaction the order of frequent items is resorted
according to the head table.
Step III: If T has a child N, then the count of N is
increased by 1, else a new node is created. The
insertion function is called recursively until P
becomes empty where P is the items list.
Eclat
Equivalence Class Transformation (Eclat)
algorithm is used to perform itemset mining by
exploring the data in vertical format. Item set
mining lets us to find frequent patterns in data[19].
This type of pattern is called association rules and
is used in many application domains. The first scan
of the database builds the TID_set of each single
item. Starting with a single item (k = 1), the
frequent (k+1)-item sets developed from a previous
k-item set can be generated according to the
Apriori property, with a depth-first computation
order. The computation is done by intersection of
the TID_sets of the frequent k-itemsets to compute
the TID_ sets of the corresponding (k+1)-itemsets.
This process repeats, until no frequent itemsets or
no candidate itemsets can be found. The significant
advantage of this method is that there is no need to
scan the database to find the support of (k + 1)-
itemsets (for k ≥ 1). This is because the TID_set of
each k-itemset carries the complete information
required for counting such support.
III. MULTIDIMENSIONAL RULE
To enhance the traditional model of
transaction this approach can associate the records
with number of attributes that helps to explain the
context where the transaction happens. These
attributes form a multi-dimensional space and
transactions can be viewed as points in space. The
dimensions can be of any kind as long as they are
useful for mining. Association rules can be
classified as single-dimensional association rules
and Multi-dimensional association rules on the
basis of dimensions. In a Single Dimensional rule
or intra-dimensional association rule, items in the
rule refer to only one dimension or Predicate [6].
Multidimensional association rules are rules that
are found from a table structured database where
each item has distinct attributes which is not found
in association rules. Each item in the rule is a pair
i.e., attributes and value of the attribute. In a
Multidimensional rule items in the rule refer to two
or more dimensions or predicates. Hence it can be
say that it has no repeated predicates.
Multidimensional association rules with
no repeated predicates are called inter-dimensional
association rules.The main difference between
association rules and multidimensional association
rules is that each item in multidimensional
association rules has distinct attributes, but there is
no such attribute in the items of association rules.
Multidimensional association rules are basically an
application of general association rule algorithms
to table-like databases. The table-like databases
may consist of condition attributes and decision
attributes. Neelu and Neeru [10] have described
that multidimensional association rules have better
accuracy than decision trees for most of the data
sets.
3.1MULTIDIMENSIONAL APPROACHES
A. Using Static Discretization
Quantitative attributes are discretized by
incorporating concept hierarchies. This process
takes place before the occurrence of mining. For
example, a concept hierarchy for age may be used
to replace the original numeric values of this
attribute by ranges, such as “1-10", “11-20", “21-
40", and so on. Here, discretization is static and
predetermined. The discretized numeric attributes,

1445 | P a g e
with their range values, can then be treated as
categorical attributes. Each range is considered as a
category. It is referred as mining multidimensional
association rules using static discretization of
quantitative attributes [16].
B. Using Dynamic Discretization
Another method to deal with numeric
attributes is to perform discretizing called “bins"
based on the distribution of the data. These bins
may be further combined during the mining
process. The discretization process is dynamic and
established so as to satisfy some mining criteria,
such as maximizing the confidence of the rules
mined. Because this strategy treats the numeric
attribute values as quantities rather than as
predefined ranges or categories, association rules
mined from this approach are also referred to as
quantitative association rules [16].
C. Using Distance Based Discretization
It considers the values that are close to
each other and grouped into the same interval. It
involves two steps, firstly intervals or clusters are
generated by the clustering technique and then
distance based association rules are generated by
searching for groups or clusters that occur
frequently together. Hence, such quantitative
association rules are also referred to as Distance-
based association rules.[16].
D. Boolean Matrix based Approach
This approach, based on Boolean matrix
[10] is used to generate the multidimensional rule
which has no repetitive predicates. A Boolean
Matrix based approach has been used to find the
frequent itemsets, the items forming a rule
originating from different dimensions. It is an
algorithm for mining multidimensional association
rules from relational databases. The algorithm
adopts Boolean relational calculus to find out
frequent predicate sets. While using algorithm for
the first time, it scans the database once and will
generate the association rules. Apriori property is
implemented in this algorithm to prune the sets. It
is not necessary to scan the database again; it uses
Boolean logical operations to generate the
association rules. It stores all data in the form of
bits, so that it reduces memory space and can be
applied to large relational databases. An algorithm
to transfer the relational database in Boolean matrix
is as follows:
The algorithm consists of the following steps:
1. Transforming the relational database in to
Boolean matrix.
2. Generating the set of frequent 1-itemsets.
3. Pruning the Boolean matrix.
4. Perform AND operation on frequent 1-
items of different dimensions to
generate 2-itemsets.
5. Repeat the process to generate more
number of item sets until the resultant
item set is generated.
E. Bit-Mask Search Algorithm
In this approach, a new array based
optimization technique called Bit stream mask and
Masked Item set Processing search is used for
mining complete frequent item sets [5]. Data
representation with additional storage of sixteen
elements in one process memory array location has
been compared with the available implementation
of familiar Apriori and FP-growth algorithms. This
technique shows improved performance of mining
frequent item sets on a number of typical datasets.
It is faster than the other Apriori Tree.
IV. QUANTITATIVE RULE
Mining association rules on both
categorical and numeric attributes is said to be
quantitative association rules. When the domain is
continuous, then the association rules are
Quantitative association rule. Numeric attributes
are always defined on a wide range of different
values. Therefore, it is difficult to work on all
possible values. A classic method to deal with
numeric attributes is to divide their domains into
intervals called Discretization.
Most of them are based on some form of
discretization labelled as partitioning, quantization
or bucketing of numeric attributes, which means
dividing the attribute domain into separate ranges.
Such pre-processing method results in the loss of
informational value of discovered rules, or often
loses significant ones. In order to find out
association rules with quantitative attributes, an
efficient method is required to reveal the
meaningful intervals.
Many researchers proposed different
methods to solve the problem of interval partition.
For quantitative attributes that are not partitioned
into intervals, the values are mapped to consecutive
integers. These mappings lead to treat a database
record as a set of (attribute, integer value) pairs,
without loss of generality. Adaptation of the
APRIORI algorithm for mining quantitative
association rules was identified soon after the
introduction of APRIORI algorithm. The necessity
for quantity in mining association rules was first
identified by Sujatha and Naveen [15]. It proposed
rules of the form x=qx⇒y=qy i.e. it associated a
single quantity ‘q’ to the antecedent and the
consequent. This was done by decomposition of
one quantitative attribute into several binary
attributes.

1446 | P a g e
However, most of the approaches are
designed for "market basket analysis" and operate
on categorical (qualitative) data. It renders them
useless for learning from many common types of
data based on numeric values. Special forms of
association rules for quantitative attributes may be
applicable.
Quantitative approach
There are only few algorithms and
methodologies to deal with quantitative
associations.
Discretization
Discretization technique [9] [10] can be
considered as the first comprehensive work on
quantitative association rules. The basic idea of the
technique is to map quantitative data to Boolean by
considering a partition of the numerical attributes
into sets of intervals. Then, an algorithm for
finding Boolean association rules can be used to
get quantitative rules.
Two main types of partitions are
introduced. A fixed partition, where the set of
intervals are disjoint and another type, where the
boundaries of intervals are overlapped with each
other. The rules are in
the form X⇒Y , where X and Y are sets of
attributes and they are either a categorical item or a
range of numerical values. The main advantage of
this technique is that it handles both categorical and
numerical data equally.
Disjoint sets suffer from Minimum
Support and Minimum Confidence whereas
overlapped sets suffer from Sharp Boundary
problem. Using intervals instead of the original
continuous data leads to loss of information. The
rules obtained will be only an approximation of the
best ones. Moreover, the lack of accuracy not only
affects the quality of the rules but also leads to
large number of unwanted rules.
Another problem is the expansion of the
attributes. The Discretization technique can be used
in any kind of data that contains any combination
of categorical and numerical attributes even on the
case of all attributes are numerical.
APACS2 (Automatic Pattern Analysis and
Classification System)
The algorithm APACS2 [9] is based on
employing both adjusted difference analysis and
discretization to find rules between two attributes.
The two attributes can be any combination of
numerical or categorical. The technique has the
ability of discovering positive and negative
association rules and does not need any user
thresholds. The rules are in the form,
(attributel) =>attributer [type]
Where attributel and attributer , may be
categorical and quantitative in any order and type
may be + or – for positive and negative rules
respectively.It has the ability of finding new
meaningful objective measure of the association
rules both positive and negative.
Statistical Inference
In some cases numerical attributes are
treated as a continuous values rather than
converting it into range of numerical values based
on statistical inference [9]. In order to make this
algorithmically manageable, two specific
association rules are defined: Categorical to
Quantitative rules with an unlimited number of
attributes, and Quantitative to Quantitative where
both sides contain a single attribute only. The
general form for the first type of association rules is
X⇒ MJ (TX) (MJ (TX) ≈MJ (D))
Where X is categorical attributes, J is a set of
numerical attributes, M is a statistical Measure, Tx
is the set of transactions defined by X, and D is the
whole database.
This rule is interesting because it brings
out the value of certain group of attributes
significantly lower than the threshold. Obviously,
the technique has a good advantage, which is the
use of numerical data as it is. Also, it introduces a
new, indeed an interesting measure which is
finding set of frequent behaviour that is different
from overall behaviour.
BAR mining Algorithm
Mining Quantitative Association Rules
(QAR) by a generic Boolean Association Rule
(BAR) mining algorithm, [17] however, is
infeasible in most cases for the following reasons.
First, QAR mining suffers from the problem of a
combinatorial explosion of attribute sets as does
BAR mining. In practice, the number of distinct
attributes in a QAR mining problem may not be as
large as that in a BAR mining problem.
However, as shown by Srikant and
Agrawal [1], it is necessary to combine the
consecutive intervals of a quantitative attribute in
order to gain sufficient support and more
meaningful intervals. This leads to another
combinatorial explosion problem if the domain of a
quantitative attribute is partitioned into n intervals;
the total number of intervals of the attribute grows
to O (n2
) afer combining the consecutive intervals.
When we join the attributes in the mining process,
the number of itemsets can become prohibitively
large if the number of intervals associated with an
attribute is large then the attributes are joined.
Effective techniques to prune the large search space
of QAR mining are necessary.

1447 | P a g e
V. ISSUES AND CHALLENGES OF
MUTIDIMENSIONAL AND
QUANTITATIVE ASSOCIATION
RULES
Association rule mining on high
dimensional data, now a days is a high topic of
research interest in many fields of data mining
tasks. There are numerous data repositories which
stores the data in different dimensions. Mining
association rules on these databases is a
challenging issue. There is a need to find
association rules on high dimensional data a more
effectively for different applications and decision
making. There has been work on quantifying the
“usefulness" or “interestingness" of a rule.
The mining of all association rules
existing in the database with respect to certain user
interesting measures. Database in this case
consisting of a data set with varying attributes or
dimensions in the average out a total set. Although
the association mining is still a topic of further
research, during recent years many algorithms for
specialized tasks have been developed. First of all,
there are the approaches that enhance the
association rules itself. E.g. quantitative association
rules, generalized association rules and to some
extent the work on sequential patterns.
The emergence of various new application
domains, such as bioinformatics and e-commerce,
emphasize the need for analyzing high dimensional
data. Many organizations have enormous amounts
of data containing valuable information for running
and building a decision making system. Extracting
the value of that data is a big challenge.
First and foremost is to understand and
analyze the large amount data for effective decision
making. For example, generally, in a gene
expression microarray data set, there could be tens
or hundreds of dimensions, each of which
corresponds to an experimental condition. In a
customer purchase behavior data set, there may be
up to hundreds of thousands of merchandizes, each
of which is mapped to a dimension. Researchers
and practitioners are very eager in analyzing these
types of data sets.
However, before analyzing the data
mining models, the researcher will analyze the
challenges of attribute selection, the curse of
dimensionality, the specification of similarity in
high dimensional space for analyzing high
dimensional data set. Association Rule Mining in
high dimensional spaces presents tremendous
difficulty in generating the rules, much better than
in predictive mining. Attribute selection is a one
which reduces the impact of high dimensional data
space at the time of generating rules.
Dimensionality Curse is a way of speaking
about lack of data separation in high dimensional
space. The complexity of many existing data
mining algorithms is exponential with respect to
the number of dimensions. With increasing
dimensionality, these algorithms soon become
computationally intractable and therefore
inapplicable in many real applications.
Multi dimensional association rule mining
carried on multi dimensional datasets which stores
the data in more than one dimensions or predicates.
Multidimensional association rules are used to
mine when the data or transactions are located in
multidimensional space, such as in a relational
database or data warehouse. Multiple dimensional
association rule mining is to discover the
correlation between different predicates or
attributes. Each attribute or predicate is called a
dimension. The data may contain different types of
data such as categorical, Boolean, numeric. The
attributes are also called as quantitative attributes.
Association rule mining on these attributes can be
carried both in static and dynamic discretization
methods.
More research work is carried under
Dynamic discretization method. In this the numeric
attributes are dynamically discretized during the
mining process so as to satisfy the mining criteria.
Predicate attributes or values in the dimensions are
converted into Boolean values so as to carry the
mining process very effectively. It is carried on the
following steps.
1) Determine the Number of Partitions for
each predicate or attribute.
2) Generation of Frequent Itemsets.
3) Generating Interesting rules using the
Minimum confidence.
Fabulous research work is carried under
Quantitative association rule ranging from low-
level predicates to multi dimensional predicate and
used for different range of applications.
VI. CONCLUSION
This research survey motivates us to find
the issues in multidimensional and quantitative
association rules. This paper is organized to
categorize the survey process as follows:
1) Rule Mining algorithm categorization
2) Multidimensional approach to generate
association rules
3) Quantitative association rules review
process
4) Existing applications which are
categorized with respect to fields which it
applied.
The Bit search is one of the greatest
techniques that provide to search all the elements
from the dataset. Bit Stream Mask Search is used to
find the frequent itemsets from the dataset.

1448 | P a g e
Multidimensional and quantitative association rule
generation is done with the help of frequent and
infrequent itemsets. The fuzzy based
multidimensional and Quantitative association rule
generation with bit mask search will definitely
identify the frequent and infrequent itemset by
reducing the scanning process significantly. This
new procedure will yield the execution time must
be reduced.
References
[1] Agrawal, R. and Srikant, R. 1994, “ Fast
algorithms for mining association rules”,
In Proc. 20th
Int. Conf. Very Large Data
Bases, VLDB, J. B. Bocca, M. Jarke, and
C. Zaniolo, Eds. Morgan Kaufmann, 487-
499.
[2] Agrawal, R., Imielinski, T., and Swami,
A. N. 1993. “Mining association rules
between sets of items in large databases”.
In Proceedings of the 1993 ACM
SIGMOD International Conference on
Management of Data, 207-216.
[3] N.Deepika* et al., (IJAEST)”
International Journal of Advanced
Engineering Sciences and Technologies”,
Vol No. 11, Issue No. 2, 253 – 257.
[4] Dimitris Kanellopoulos et.al., GESTS,
“International Transactions on Computer
Science and Engineering “, Vol.32 (1),
2006, pp. 71-82, Association Rules
Mining: A Recent Overview.
[5] E.Ramaraj and N.Venkatesan,” Bit Stream
Mask-search algorithm in Frequent Item
set Mining”, in European Journal of
scientific research. Vol.27 No.2 (2009),
pp.286-297.
[6] Han, J. 1995. Mining knowledge at
multiple concept levels. In CIKM. 19-24.
[7] Hegland, M., “Algorithms for Association
Rules, Lecture Notes in Computer
Science”, Volume 2600, Jan 2003, Pages
226 – 234.
[8] Milan Kumari et.al ., “Comparative Study
of Data Mining Classification Methods in
Cardiovascular Disease Prediction “ in “
IJCST Vol. 2, Iss ue 2, June 2011 I S S N :
2 2 2 9 - 4 3 3 3 ( P r i n t ) | I S S N : 0 9 7
6 - 8 4 9 1 (On l i n e ),.
[9] Mohamed Amin Eldesouky Elmetwaly
Elkhoul Further remarks on Quantitative
Association Rules.
[10] An Algorithm for Mining
Multidimensional Association Rules using
Boolean Matrix by Neelu Khare , Neeru
Adlakha and K.R.Pardasani.
[11] Mining Multi-Dimensional Quantitative
Associations Michaª Okoniewski, .ukasz
Gancarz, Piotr Gawryiak
[12] An Efficient Algorithm for Mining
Multilevel Association Rule Based on
Pincer Search Dr. Pratima Gautam, Rahul
Shukla.
[13] Mining Quantitative Association Rules in
Large Relational Tables by Ramakrishnan
Srikant and Rakesh Agarwal.
[14] Association Rule Mining :A Survey
Qiankun Zhoa and Sourav Bhowmick
[15] Quantitative Association Rule Mining on
Weighted Transactional Data D. Sujatha
and Naveen CH International Journal of
Information and Education Technology,
Vol. 1, No. 3, August 2011.
[16] International Journal of Emerging Science
and Engineering (IJESE) ISSN: 2319–
6378, Volume- 1, Issue-4,
February 2013 Survey on
Multidimensional and Conditional Hybrid
Dimensional Association Rule Mining.
[17] An Information – Theoretic Approach to
Quantitative Association Rule Mining by
Yiping Ke James and Cheng Wilfred Ng.
[18] An Algorithm for Mining
multidimensional Fuzzy Association
Rules by Neelu Khare , Neeru
Adlakha and K.R.Pardasani
[19] Frequent pattern mining:current status
and future directions.Jiawei
Han,Hong Cheng,Dong Xin,Xifeng
Yan.

Hu3414421448

More Related Content

What's hot (17)

Viewers also liked (20)

Similar to Hu3414421448 (20)

Recently uploaded (20)

Hu3414421448